From 0eebfa7601437ed8b2a122cd5e927edf5cb006a8 Mon Sep 17 00:00:00 2001 From: "docling-java-ops[bot]" <245010719+docling-java-ops[bot]@users.noreply.github.com> Date: Mon, 24 Aug 2026 04:37:25 +0000 Subject: [PATCH] docs: Update docling-serve compatibility table (from Version tests run # 65) Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> --- .../docling-serve/serve-compatibility.md | 3541 +++++++++-------- 1 file changed, 1862 insertions(+), 1679 deletions(-) diff --git a/docs/src/doc/docs/includes/docling-serve/serve-compatibility.md b/docs/src/doc/docs/includes/docling-serve/serve-compatibility.md index 7034c536..33ca4758 100644 --- a/docs/src/doc/docs/includes/docling-serve/serve-compatibility.md +++ b/docs/src/doc/docs/includes/docling-serve/serve-compatibility.md @@ -1,9 +1,10 @@ -# Results for ghcr.io/docling-project/docling-serve as of 2026-08-17T13:10:37.153488382Z +# Results for ghcr.io/docling-project/docling-serve as of 2026-08-24T04:37:21.526667782Z Here are the results: | Tag | Result | Details | | --- | ------ | ------- | +| v1.31.0 | ✅ SUCCESS | [Click for run details](#v1.31.0-details) | | v1.30.0 | ✅ SUCCESS | [Click for run details](#v1.30.0-details) | | v1.29.0 | ✅ SUCCESS | [Click for run details](#v1.29.0-details) | | v1.28.0 | ✅ SUCCESS | [Click for run details](#v1.28.0-details) | @@ -51,6 +52,188 @@ Here are the results: ## Details +### ghcr.io/docling-project/docling-serve:v1.31.0 + +
+Click to expand + +#### Message + +
+Click to collapse + +~~~markdown +Tag v1.31.0 is ok +~~~ + +
+ + +#### Docling server logs + +
+click to expand + +``` +Starting production server 🚀 + +Server started at http://0.0.0.0:5001 +Documentation at http://0.0.0.0:5001/docs +Scalar docs at http://0.0.0.0:5001/scalar + +Logs: +INFO: 04:37:04 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:37:04 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:37:04 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' +INFO: 04:37:05 - docling_jobkit.connectors.plugins.defaults - Connector 'SharePointSourceProcessor' skipped — optional dependency not installed (No module named 'office365'). Install the matching extra to enable it. +INFO: 04:37:05 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' +INFO: 04:37:05 - docling_jobkit.connectors.plugins.defaults - Connector 'OpenSearchTargetProcessor' skipped — optional dependency not installed (No module named 'opensearchpy'). Install the matching extra to enable it. +INFO: 04:37:05 - docling_jobkit.connectors.plugins.defaults - Connector 'AstraDBTargetProcessor' skipped — optional dependency not installed (No module named 'astrapy'). Install the matching extra to enable it. +INFO: 04:37:05 - docling_jobkit.connectors.plugins.defaults - Connector 'SharePointTargetProcessor' skipped — optional dependency not installed (No module named 'office365'). Install the matching extra to enable it. +INFO: 04:37:05 - docling_jobkit.connectors.plugins.defaults - Connector 'KafkaTargetProcessor' skipped — optional dependency not installed (No module named 'confluent_kafka'). Install the matching extra to enable it. +INFO: 04:37:05 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:37:05 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:37:05 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/metrics$,/health$,/livez$,/ready$,/healthz$,/readyz$) +INFO: 04:37:05 - uvicorn.error - Started server process [1] +INFO: 04:37:05 - uvicorn.error - Waiting for application startup. +INFO: 04:37:08 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:37:08 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:37:08 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:37:08 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:37:08 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:37:08 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 9d06178a9f1c8aeb9889d1e46440f3fa +INFO: 04:37:08 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:37:08 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +INFO: 04:37:08 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +2026-08-24 04:37:08.927188507 [W:onnxruntime:Default, device_discovery.cc:146 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:37:09 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:37:09,158 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:37:09,159 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:37:09,214 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:37:09,215 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:37:09,257 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:37:09,258 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_rec_small.onnx +INFO: 04:37:09 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:37:09 - docling.models.inference_engines.object_detection.transformers_engine - Initializing Transformers object-detection engine +INFO: 04:37:09 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:37:11 - docling.document_converter - Going to convert document batch... +INFO: 04:37:11 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash a1d02772fb798c5f3e06aa82e0d32e7d +INFO: 04:37:11 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:37:12 - docling.document_converter - Finished converting document file in 0.40 sec. +/opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:641: RuntimeWarning: Mean of empty slice + np.nanmean( +/opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered + return _nanquantile_unchecked( +WARNING: 04:37:12 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:37:12 - docling_jobkit.convert.results - Processed 1 docs in 0.47 seconds. +INFO: 04:37:12 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job dc658ce0-e5aa-4e1d-aa92-2bb35dde963e in 0.47 seconds +INFO: 04:37:12 - uvicorn.access - 172.17.0.1:60420 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:37:12 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:37:12 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:37:12 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:37:12 - docling_serve.app - [TENANT_ID] Task 0c5fa041-e811-411a-ad3b-533591cbfc11 created with tenant_id='default' +INFO: 04:37:12 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 0c5fa041-e811-411a-ad3b-533591cbfc11 +INFO: 04:37:12 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:37:12 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash a1d02772fb798c5f3e06aa82e0d32e7d +INFO: 04:37:12 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:37:12 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:37:12,781 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:37:12,781 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:37:12,834 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:37:12,834 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:37:12,881 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:37:12,881 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_rec_small.onnx +INFO: 04:37:12 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:37:12 - docling.models.inference_engines.object_detection.transformers_engine - Initializing Transformers object-detection engine +INFO: 04:37:12 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:37:13 - docling.document_converter - Going to convert document batch... +INFO: 04:37:13 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash a1d02772fb798c5f3e06aa82e0d32e7d +INFO: 04:37:13 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:37:13 - docling.document_converter - Finished converting document file in 0.16 sec. +/opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:641: RuntimeWarning: Mean of empty slice + np.nanmean( +/opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered + return _nanquantile_unchecked( +INFO: 04:37:13 - docling_jobkit.convert.results - Processed 1 docs in 0.16 seconds. +INFO: 04:37:13 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 0c5fa041-e811-411a-ad3b-533591cbfc11 in 0.16 seconds +INFO: 04:37:14 - uvicorn.access - 172.17.0.1:60420 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:37:14 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:37:14 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:37:14 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:37:14 - docling_serve.app - [TENANT_ID] Task f36274fd-42a8-45f9-98af-1e5bcfcb3f45 created with tenant_id='default' +INFO: 04:37:14 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task f36274fd-42a8-45f9-98af-1e5bcfcb3f45 +INFO: 04:37:14 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:37:14 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 21c1cdafbd5b48f9b0cfcb806e9e9d57 +INFO: 04:37:14 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:37:14 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:37:14,793 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:37:14,793 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:37:14,842 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:37:14,842 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:37:14,897 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:37:14,897 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_rec_small.onnx +INFO: 04:37:14 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:37:14 - docling.models.inference_engines.object_detection.transformers_engine - Initializing Transformers object-detection engine +INFO: 04:37:14 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:37:15 - docling.document_converter - Going to convert document batch... +INFO: 04:37:15 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 21c1cdafbd5b48f9b0cfcb806e9e9d57 +INFO: 04:37:15 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:37:16 - docling.document_converter - Finished converting document file in 0.36 sec. +/opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:641: RuntimeWarning: Mean of empty slice + np.nanmean( +/opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered + return _nanquantile_unchecked( +INFO: 04:37:16 - docling_jobkit.convert.results - Processed 1 docs in 0.39 seconds. +INFO: 04:37:16 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job f36274fd-42a8-45f9-98af-1e5bcfcb3f45 in 0.39 seconds +INFO: 04:37:16 - uvicorn.access - 172.17.0.1:60420 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:37:16 - uvicorn.access - 172.17.0.1:60420 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:37:16 - uvicorn.access - 172.17.0.1:60420 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 + +``` + +
+ +
+ ### ghcr.io/docling-project/docling-serve:v1.30.0
@@ -81,151 +264,151 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -INFO: 13:10:18 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:10:18 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] -INFO: 13:10:18 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' -INFO: 13:10:19 - docling_jobkit.connectors.plugins.defaults - Connector 'SharePointSourceProcessor' skipped — optional dependency not installed (No module named 'office365'). Install the matching extra to enable it. -INFO: 13:10:19 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' -INFO: 13:10:19 - docling_jobkit.connectors.plugins.defaults - Connector 'OpenSearchTargetProcessor' skipped — optional dependency not installed (No module named 'opensearchpy'). Install the matching extra to enable it. -INFO: 13:10:19 - docling_jobkit.connectors.plugins.defaults - Connector 'AstraDBTargetProcessor' skipped — optional dependency not installed (No module named 'astrapy'). Install the matching extra to enable it. -INFO: 13:10:19 - docling_jobkit.connectors.plugins.defaults - Connector 'SharePointTargetProcessor' skipped — optional dependency not installed (No module named 'office365'). Install the matching extra to enable it. -INFO: 13:10:19 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics -INFO: 13:10:19 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export -INFO: 13:10:19 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/ready$,/readyz$,/healthz$,/health$,/livez$,/metrics$) -INFO: 13:10:19 - uvicorn.error - Started server process [1] -INFO: 13:10:19 - uvicorn.error - Waiting for application startup. -INFO: 13:10:22 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:10:22 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] -INFO: 13:10:22 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:10:22 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] -INFO: 13:10:22 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:10:22 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash e2db689ab35ca52a08b5b7fa25de8cc8 -INFO: 13:10:22 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:10:22 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] -INFO: 13:10:22 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -2026-08-17 13:10:22.614978911 [W:onnxruntime:Default, device_discovery.cc:134 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -INFO: 13:10:22 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:10:22,845 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:10:22,847 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:10:22,903 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:10:22,903 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:10:22,945 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:10:22,945 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_rec_small.onnx -INFO: 13:10:23 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:10:23 - docling.models.inference_engines.object_detection.transformers_engine - Initializing Transformers object-detection engine -INFO: 13:10:23 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +INFO: 04:35:54 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:35:54 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:35:54 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' +INFO: 04:35:55 - docling_jobkit.connectors.plugins.defaults - Connector 'SharePointSourceProcessor' skipped — optional dependency not installed (No module named 'office365'). Install the matching extra to enable it. +INFO: 04:35:55 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' +INFO: 04:35:55 - docling_jobkit.connectors.plugins.defaults - Connector 'OpenSearchTargetProcessor' skipped — optional dependency not installed (No module named 'opensearchpy'). Install the matching extra to enable it. +INFO: 04:35:55 - docling_jobkit.connectors.plugins.defaults - Connector 'AstraDBTargetProcessor' skipped — optional dependency not installed (No module named 'astrapy'). Install the matching extra to enable it. +INFO: 04:35:55 - docling_jobkit.connectors.plugins.defaults - Connector 'SharePointTargetProcessor' skipped — optional dependency not installed (No module named 'office365'). Install the matching extra to enable it. +INFO: 04:35:55 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:35:55 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:35:55 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/metrics$,/readyz$,/healthz$,/ready$,/health$,/livez$) +INFO: 04:35:55 - uvicorn.error - Started server process [1] +INFO: 04:35:55 - uvicorn.error - Waiting for application startup. +INFO: 04:35:58 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:35:59 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:35:59 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:35:59 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:35:59 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:35:59 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash e2db689ab35ca52a08b5b7fa25de8cc8 +INFO: 04:35:59 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:35:59 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +INFO: 04:35:59 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +2026-08-24 04:35:59.672906055 [W:onnxruntime:Default, device_discovery.cc:134 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:35:59 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:35:59,979 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:35:59,980 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:36:00,082 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:36:00,083 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:36:00,174 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:36:00,175 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_rec_small.onnx +INFO: 04:36:00 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:36:00 - docling.models.inference_engines.object_detection.transformers_engine - Initializing Transformers object-detection engine +INFO: 04:36:00 - docling.utils.accelerator_utils - Accelerator device: 'cpu' [transformers] `torch_dtype` is deprecated! Use `dtype` instead! - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:10:27 - docling.document_converter - Going to convert document batch... -INFO: 13:10:27 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 5cd3e3811d2e4d7d996fcfcbbe327b1d -INFO: 13:10:27 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:10:27 - docling.document_converter - Finished converting document file in 0.40 sec. + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:36:05 - docling.document_converter - Going to convert document batch... +INFO: 04:36:05 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 5cd3e3811d2e4d7d996fcfcbbe327b1d +INFO: 04:36:05 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:36:06 - docling.document_converter - Finished converting document file in 0.50 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:635: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -WARNING: 13:10:27 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. -INFO: 13:10:27 - docling_jobkit.convert.results - Processed 1 docs in 0.44 seconds. -INFO: 13:10:27 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 54e7ee3d-bd53-403f-8212-8cfe5fda61c3 in 0.44 seconds -INFO: 13:10:28 - uvicorn.access - 172.17.0.1:40808 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:10:28 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:10:28 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:10:28 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:10:28 - docling_serve.app - [TENANT_ID] Task ba972c3c-e35a-472d-a553-94609d1ba3e2 created with tenant_id='default' -INFO: 13:10:28 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task ba972c3c-e35a-472d-a553-94609d1ba3e2 -INFO: 13:10:28 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:10:28 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 5cd3e3811d2e4d7d996fcfcbbe327b1d -INFO: 13:10:28 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -INFO: 13:10:28 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:10:28,205 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:10:28,205 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:10:28,257 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:10:28,257 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:10:28,321 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:10:28,321 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_rec_small.onnx -INFO: 13:10:28 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:10:28 - docling.models.inference_engines.object_detection.transformers_engine - Initializing Transformers object-detection engine -INFO: 13:10:28 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:10:29 - docling.document_converter - Going to convert document batch... -INFO: 13:10:29 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 5cd3e3811d2e4d7d996fcfcbbe327b1d -INFO: 13:10:29 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:10:29 - docling.document_converter - Finished converting document file in 0.39 sec. +WARNING: 04:36:06 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:36:06 - docling_jobkit.convert.results - Processed 1 docs in 0.61 seconds. +INFO: 04:36:06 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 3e38ca39-d462-4e95-a1f1-2c1a7948b32e in 0.61 seconds +INFO: 04:36:08 - uvicorn.access - 172.17.0.1:54782 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:36:08 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:36:08 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:36:08 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:36:08 - docling_serve.app - [TENANT_ID] Task d2cd02b3-5b25-4952-83bc-e3290326e161 created with tenant_id='default' +INFO: 04:36:08 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task d2cd02b3-5b25-4952-83bc-e3290326e161 +INFO: 04:36:08 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:36:08 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 5cd3e3811d2e4d7d996fcfcbbe327b1d +INFO: 04:36:08 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:36:08 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:36:08,342 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:36:08,342 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:36:08,398 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:36:08,399 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:36:08,505 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:36:08,506 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_rec_small.onnx +INFO: 04:36:08 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:36:08 - docling.models.inference_engines.object_detection.transformers_engine - Initializing Transformers object-detection engine +INFO: 04:36:08 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:36:10 - docling.document_converter - Going to convert document batch... +INFO: 04:36:10 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 5cd3e3811d2e4d7d996fcfcbbe327b1d +INFO: 04:36:10 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:36:10 - docling.document_converter - Finished converting document file in 0.98 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:635: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -INFO: 13:10:29 - docling_jobkit.convert.results - Processed 1 docs in 0.39 seconds. -INFO: 13:10:29 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job ba972c3c-e35a-472d-a553-94609d1ba3e2 in 0.39 seconds -INFO: 13:10:30 - uvicorn.access - 172.17.0.1:40808 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:10:30 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:10:30 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:10:30 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:10:30 - docling_serve.app - [TENANT_ID] Task a45832fa-2409-4a1b-8c7c-32ebe272b41b created with tenant_id='default' -INFO: 13:10:30 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task a45832fa-2409-4a1b-8c7c-32ebe272b41b -INFO: 13:10:30 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:10:30 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 9df8df3fc03c33cf6e8bd583e67dce64 -INFO: 13:10:30 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -INFO: 13:10:30 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:10:30,215 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:10:30,215 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:10:30,266 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:10:30,266 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:10:30,309 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:10:30,309 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_rec_small.onnx -INFO: 13:10:30 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:10:30 - docling.models.inference_engines.object_detection.transformers_engine - Initializing Transformers object-detection engine -INFO: 13:10:30 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:10:31 - docling.document_converter - Going to convert document batch... -INFO: 13:10:31 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 9df8df3fc03c33cf6e8bd583e67dce64 -INFO: 13:10:31 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:10:31 - docling.document_converter - Finished converting document file in 0.38 sec. +INFO: 04:36:10 - docling_jobkit.convert.results - Processed 1 docs in 0.98 seconds. +INFO: 04:36:10 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job d2cd02b3-5b25-4952-83bc-e3290326e161 in 0.98 seconds +INFO: 04:36:12 - uvicorn.access - 172.17.0.1:54782 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:36:12 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:36:12 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:36:12 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:36:12 - docling_serve.app - [TENANT_ID] Task 863fb628-b6e3-43c3-9022-5c0c85486cbc created with tenant_id='default' +INFO: 04:36:12 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 863fb628-b6e3-43c3-9022-5c0c85486cbc +INFO: 04:36:12 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:36:12 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 9df8df3fc03c33cf6e8bd583e67dce64 +INFO: 04:36:12 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:36:12 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:36:12,665 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:36:12,666 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:36:12,743 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:36:12,743 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:36:12,817 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:36:12,817 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/PP-OCRv6_rec_small.onnx +INFO: 04:36:12 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:36:12 - docling.models.inference_engines.object_detection.transformers_engine - Initializing Transformers object-detection engine +INFO: 04:36:12 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:36:14 - docling.document_converter - Going to convert document batch... +INFO: 04:36:14 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 9df8df3fc03c33cf6e8bd583e67dce64 +INFO: 04:36:14 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:36:14 - docling.document_converter - Finished converting document file in 0.46 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:635: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -INFO: 13:10:31 - docling_jobkit.convert.results - Processed 1 docs in 0.40 seconds. -INFO: 13:10:31 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job a45832fa-2409-4a1b-8c7c-32ebe272b41b in 0.40 seconds -INFO: 13:10:32 - uvicorn.access - 172.17.0.1:40808 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:10:32 - uvicorn.access - 172.17.0.1:40808 - "GET /v1/clear/converters HTTP/1.1" 200 -INFO: 13:10:32 - uvicorn.access - 172.17.0.1:40808 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 +INFO: 04:36:14 - docling_jobkit.convert.results - Processed 1 docs in 0.52 seconds. +INFO: 04:36:14 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 863fb628-b6e3-43c3-9022-5c0c85486cbc in 0.52 seconds +INFO: 04:36:16 - uvicorn.access - 172.17.0.1:54782 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:36:16 - uvicorn.access - 172.17.0.1:54782 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:36:16 - uvicorn.access - 172.17.0.1:54782 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 ``` @@ -263,139 +446,139 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -INFO: 13:08:44 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:08:44 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] -INFO: 13:08:44 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' -INFO: 13:08:45 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' -INFO: 13:08:45 - docling_jobkit.connectors.plugins.defaults - Connector 'OpenSearchTargetProcessor' skipped — optional dependency not installed (No module named 'opensearchpy'). Install the matching extra to enable it. -INFO: 13:08:45 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics -INFO: 13:08:45 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export -INFO: 13:08:45 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/healthz$,/health$,/livez$,/readyz$,/metrics$,/ready$) -INFO: 13:08:45 - uvicorn.error - Started server process [1] -INFO: 13:08:45 - uvicorn.error - Waiting for application startup. -INFO: 13:08:48 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:08:48 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] -INFO: 13:08:49 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:08:49 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] -INFO: 13:08:49 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:08:49 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash caff667a02b7f037838d32d34acfe0b0 -INFO: 13:08:49 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:08:49 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] -INFO: 13:08:49 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -2026-08-17 13:08:49.200480219 [W:onnxruntime:Default, device_discovery.cc:134 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -INFO: 13:08:49 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:08:49,470 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:08:49,472 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:08:49,528 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:08:49,528 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:08:49,572 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:08:49,572 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx -INFO: 13:08:49 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:08:49 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:08:52 - docling.document_converter - Going to convert document batch... -INFO: 13:08:52 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:08:52 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:08:52 - docling.document_converter - Finished converting document file in 0.42 sec. +INFO: 04:34:36 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:34:36 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:34:36 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' +INFO: 04:34:37 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' +INFO: 04:34:37 - docling_jobkit.connectors.plugins.defaults - Connector 'OpenSearchTargetProcessor' skipped — optional dependency not installed (No module named 'opensearchpy'). Install the matching extra to enable it. +INFO: 04:34:37 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:34:37 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:34:37 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/health$,/metrics$,/healthz$,/ready$,/livez$,/readyz$) +INFO: 04:34:37 - uvicorn.error - Started server process [1] +INFO: 04:34:37 - uvicorn.error - Waiting for application startup. +INFO: 04:34:43 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:34:44 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:34:44 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:34:44 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:34:44 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:34:44 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash caff667a02b7f037838d32d34acfe0b0 +INFO: 04:34:45 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:34:45 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +INFO: 04:34:45 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +2026-08-24 04:34:45.062942663 [W:onnxruntime:Default, device_discovery.cc:134 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:34:45 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:34:45,848 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:34:45,851 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:34:45,995 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:34:45,995 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:34:46,058 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:34:46,059 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx +INFO: 04:34:46 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:34:46 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:34:53 - docling.document_converter - Going to convert document batch... +INFO: 04:34:53 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:34:53 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:34:53 - docling.document_converter - Finished converting document file in 0.59 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:632: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -WARNING: 13:08:53 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. -INFO: 13:08:53 - docling_jobkit.convert.results - Processed 1 docs in 0.46 seconds. -INFO: 13:08:53 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 1702ef89-de0f-4d14-995e-e8a937841be1 in 0.46 seconds -INFO: 13:08:53 - uvicorn.access - 172.17.0.1:55020 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:08:53 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:08:53 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:08:53 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:08:53 - docling_serve.app - [TENANT_ID] Task f71f5798-c675-48d9-b86e-00dfb5aa0139 created with tenant_id='default' -INFO: 13:08:53 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task f71f5798-c675-48d9-b86e-00dfb5aa0139 -INFO: 13:08:53 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:08:53 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash ab12ef348c8fc273bf2be7a49773b303 -INFO: 13:08:53 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -INFO: 13:08:53 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:08:53,319 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:08:53,319 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:08:53,375 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:08:53,375 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:08:53,429 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:08:53,429 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx -INFO: 13:08:53 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:08:53 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:08:54 - docling.document_converter - Going to convert document batch... -INFO: 13:08:54 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:08:54 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:08:54 - docling.document_converter - Finished converting document file in 0.39 sec. +WARNING: 04:34:53 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:34:53 - docling_jobkit.convert.results - Processed 1 docs in 0.70 seconds. +INFO: 04:34:53 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 682b9685-806c-45c8-a9fe-d25e8e98d4ab in 0.70 seconds +INFO: 04:34:54 - uvicorn.access - 172.17.0.1:42504 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:34:54 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:34:54 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:34:54 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:34:54 - docling_serve.app - [TENANT_ID] Task 7d8e5113-45ac-4cf5-b285-0d4cdf102889 created with tenant_id='default' +INFO: 04:34:54 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 7d8e5113-45ac-4cf5-b285-0d4cdf102889 +INFO: 04:34:54 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:34:55 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash ab12ef348c8fc273bf2be7a49773b303 +INFO: 04:34:55 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:34:55 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:34:55,179 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:34:55,179 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:34:55,242 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:34:55,243 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:34:55,322 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:34:55,323 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx +INFO: 04:34:55 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:34:55 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:34:56 - docling.document_converter - Going to convert document batch... +INFO: 04:34:56 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:34:56 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:34:56 - docling.document_converter - Finished converting document file in 0.48 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:632: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -INFO: 13:08:54 - docling_jobkit.convert.results - Processed 1 docs in 0.39 seconds. -INFO: 13:08:54 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job f71f5798-c675-48d9-b86e-00dfb5aa0139 in 0.39 seconds -INFO: 13:08:55 - uvicorn.access - 172.17.0.1:55020 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:08:55 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:08:55 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:08:55 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:08:55 - docling_serve.app - [TENANT_ID] Task c66164aa-4710-4523-84d0-f68cb62185c3 created with tenant_id='default' -INFO: 13:08:55 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task c66164aa-4710-4523-84d0-f68cb62185c3 -INFO: 13:08:55 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:08:55 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash b3849eabdc7ecae5f3d54428a0162342 -INFO: 13:08:55 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -INFO: 13:08:55 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:08:55,349 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:08:55,350 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:08:55,400 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:08:55,401 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:08:55,453 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:08:55,453 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx -INFO: 13:08:55 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:08:55 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:08:56 - docling.document_converter - Going to convert document batch... -INFO: 13:08:56 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:08:56 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:08:56 - docling.document_converter - Finished converting document file in 0.23 sec. +INFO: 04:34:56 - docling_jobkit.convert.results - Processed 1 docs in 0.48 seconds. +INFO: 04:34:56 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 7d8e5113-45ac-4cf5-b285-0d4cdf102889 in 0.48 seconds +INFO: 04:34:56 - uvicorn.access - 172.17.0.1:42504 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:34:56 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:34:56 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:34:56 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:34:56 - docling_serve.app - [TENANT_ID] Task a5e72965-82d7-4342-b438-61c21957d2f9 created with tenant_id='default' +INFO: 04:34:56 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task a5e72965-82d7-4342-b438-61c21957d2f9 +INFO: 04:34:56 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:34:57 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash b3849eabdc7ecae5f3d54428a0162342 +INFO: 04:34:57 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:34:57 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:34:57,220 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:34:57,221 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:34:57,312 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:34:57,313 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:34:57,380 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:34:57,380 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx +INFO: 04:34:57 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:34:57 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:34:59 - docling.document_converter - Going to convert document batch... +INFO: 04:34:59 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:34:59 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:34:59 - docling.document_converter - Finished converting document file in 0.46 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:632: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -INFO: 13:08:56 - docling_jobkit.convert.results - Processed 1 docs in 0.26 seconds. -INFO: 13:08:56 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job c66164aa-4710-4523-84d0-f68cb62185c3 in 0.26 seconds -INFO: 13:08:57 - uvicorn.access - 172.17.0.1:55020 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:08:57 - uvicorn.access - 172.17.0.1:55020 - "GET /v1/clear/converters HTTP/1.1" 200 -INFO: 13:08:57 - uvicorn.access - 172.17.0.1:55020 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 +INFO: 04:34:59 - docling_jobkit.convert.results - Processed 1 docs in 0.51 seconds. +INFO: 04:34:59 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job a5e72965-82d7-4342-b438-61c21957d2f9 in 0.51 seconds +INFO: 04:35:00 - uvicorn.access - 172.17.0.1:42504 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:35:01 - uvicorn.access - 172.17.0.1:42504 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:35:01 - uvicorn.access - 172.17.0.1:42504 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 ``` @@ -433,138 +616,138 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -INFO: 13:07:02 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:07:02 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] -INFO: 13:07:02 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' -INFO: 13:07:03 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' -INFO: 13:07:03 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics -INFO: 13:07:03 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export -INFO: 13:07:03 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/livez$,/readyz$,/metrics$,/healthz$,/ready$,/health$) -INFO: 13:07:03 - uvicorn.error - Started server process [1] -INFO: 13:07:03 - uvicorn.error - Waiting for application startup. -INFO: 13:07:06 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:07:06 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] -INFO: 13:07:06 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:07:06 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] -INFO: 13:07:06 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:07:07 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 118619a1d3ed3b201b814dac997d6742 -INFO: 13:07:07 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:07:07 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] -INFO: 13:07:07 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -2026-08-17 13:07:07.104341798 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -INFO: 13:07:07 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:07:07,390 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:07:07,392 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:07:07,454 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:07:07,455 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:07:07,503 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:07:07,503 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx -INFO: 13:07:07 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:07:07 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:07:10 - docling.document_converter - Going to convert document batch... -INFO: 13:07:10 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:07:10 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:07:10 - docling.document_converter - Finished converting document file in 0.41 sec. +INFO: 04:33:02 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:33:02 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:33:02 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' +INFO: 04:33:03 - docling_jobkit.connectors.connector_factory - Loading connector plugin 'docling_jobkit_defaults' +INFO: 04:33:03 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:33:03 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:33:03 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/ready$,/metrics$,/healthz$,/health$,/readyz$,/livez$) +INFO: 04:33:03 - uvicorn.error - Started server process [1] +INFO: 04:33:03 - uvicorn.error - Waiting for application startup. +INFO: 04:33:06 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:33:06 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:33:06 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:33:06 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:33:06 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:33:06 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 118619a1d3ed3b201b814dac997d6742 +INFO: 04:33:06 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:33:06 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +INFO: 04:33:06 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +2026-08-24 04:33:06.934194534 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:33:07 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:33:07,193 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:33:07,194 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:33:07,253 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:33:07,254 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:33:07,303 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:33:07,303 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx +INFO: 04:33:07 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:33:07 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:33:11 - docling.document_converter - Going to convert document batch... +INFO: 04:33:11 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:33:11 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:33:11 - docling.document_converter - Finished converting document file in 0.50 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:611: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -WARNING: 13:07:10 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. -INFO: 13:07:10 - docling_jobkit.convert.results - Processed 1 docs in 0.45 seconds. -INFO: 13:07:10 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 7ad07b55-a053-4fbf-86fc-2c4326ca1712 in 0.45 seconds -INFO: 13:07:11 - uvicorn.access - 172.17.0.1:48178 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:07:11 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:07:11 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:07:11 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:07:11 - docling_serve.app - [TENANT_ID] Task 3b7b6a45-dbe3-46b8-89de-ff0d0c2b68bb created with tenant_id='default' -INFO: 13:07:11 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 3b7b6a45-dbe3-46b8-89de-ff0d0c2b68bb -INFO: 13:07:11 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:07:11 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 83f6e7119a1499def48d21cd30b03869 -INFO: 13:07:11 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -INFO: 13:07:11 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:07:11,245 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:07:11,245 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:07:11,302 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:07:11,302 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:07:11,351 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:07:11,351 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx -INFO: 13:07:11 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:07:11 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:07:12 - docling.document_converter - Going to convert document batch... -INFO: 13:07:12 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:07:12 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:07:12 - docling.document_converter - Finished converting document file in 0.24 sec. +WARNING: 04:33:11 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:33:11 - docling_jobkit.convert.results - Processed 1 docs in 0.61 seconds. +INFO: 04:33:11 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job f217b197-b102-45cf-9777-fdcf5ee5a837 in 0.61 seconds +INFO: 04:33:13 - uvicorn.access - 172.17.0.1:34408 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:33:13 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:33:13 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:33:13 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:33:13 - docling_serve.app - [TENANT_ID] Task 3dde82c9-7e60-47bc-a292-6cc0fdd3ce1f created with tenant_id='default' +INFO: 04:33:13 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 3dde82c9-7e60-47bc-a292-6cc0fdd3ce1f +INFO: 04:33:13 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:33:13 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 83f6e7119a1499def48d21cd30b03869 +INFO: 04:33:13 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:33:13 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:33:13,255 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:33:13,255 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:33:13,334 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:33:13,335 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:33:13,384 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:33:13,385 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx +INFO: 04:33:13 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:33:13 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:33:14 - docling.document_converter - Going to convert document batch... +INFO: 04:33:14 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:33:14 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:33:14 - docling.document_converter - Finished converting document file in 0.44 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:611: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -INFO: 13:07:12 - docling_jobkit.convert.results - Processed 1 docs in 0.24 seconds. -INFO: 13:07:12 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 3b7b6a45-dbe3-46b8-89de-ff0d0c2b68bb in 0.24 seconds -INFO: 13:07:13 - uvicorn.access - 172.17.0.1:48178 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:07:13 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:07:13 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:07:13 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:07:13 - docling_serve.app - [TENANT_ID] Task 80b2e282-7b86-4df5-ae00-f6887a1effb4 created with tenant_id='default' -INFO: 13:07:13 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 80b2e282-7b86-4df5-ae00-f6887a1effb4 -INFO: 13:07:13 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:07:13 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 528ff53b5cc66cd5f7f61ef2cd72a8fc -INFO: 13:07:13 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -INFO: 13:07:13 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:07:13,257 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:07:13,257 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:07:13,315 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:07:13,316 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:07:13,378 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:07:13,379 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx -INFO: 13:07:13 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:07:13 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:07:14 - docling.document_converter - Going to convert document batch... -INFO: 13:07:14 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:07:14 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:07:14 - docling.document_converter - Finished converting document file in 0.42 sec. +INFO: 04:33:14 - docling_jobkit.convert.results - Processed 1 docs in 0.44 seconds. +INFO: 04:33:14 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 3dde82c9-7e60-47bc-a292-6cc0fdd3ce1f in 0.44 seconds +INFO: 04:33:15 - uvicorn.access - 172.17.0.1:34408 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:33:15 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:33:15 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:33:15 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:33:15 - docling_serve.app - [TENANT_ID] Task 39b932cc-82fe-4766-960f-c7e3560cd00f created with tenant_id='default' +INFO: 04:33:15 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 39b932cc-82fe-4766-960f-c7e3560cd00f +INFO: 04:33:15 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:33:15 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 528ff53b5cc66cd5f7f61ef2cd72a8fc +INFO: 04:33:15 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:33:15 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:33:15,326 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:33:15,326 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:33:15,399 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:33:15,400 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:33:15,469 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:33:15,469 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx +INFO: 04:33:15 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:33:15 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:33:16 - docling.document_converter - Going to convert document batch... +INFO: 04:33:16 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:33:16 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:33:16 - docling.document_converter - Finished converting document file in 0.44 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:611: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -INFO: 13:07:14 - docling_jobkit.convert.results - Processed 1 docs in 0.44 seconds. -INFO: 13:07:14 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 80b2e282-7b86-4df5-ae00-f6887a1effb4 in 0.44 seconds -INFO: 13:07:15 - uvicorn.access - 172.17.0.1:48178 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:07:15 - uvicorn.access - 172.17.0.1:48178 - "GET /v1/clear/converters HTTP/1.1" 200 -INFO: 13:07:15 - uvicorn.access - 172.17.0.1:48178 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 +INFO: 04:33:16 - docling_jobkit.convert.results - Processed 1 docs in 0.49 seconds. +INFO: 04:33:16 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 39b932cc-82fe-4766-960f-c7e3560cd00f in 0.49 seconds +INFO: 04:33:17 - uvicorn.access - 172.17.0.1:34408 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:33:17 - uvicorn.access - 172.17.0.1:34408 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:33:17 - uvicorn.access - 172.17.0.1:34408 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 ``` @@ -602,136 +785,136 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -INFO: 13:05:14 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:05:14 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] -INFO: 13:05:14 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics -INFO: 13:05:14 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export -INFO: 13:05:14 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/ready$,/health$,/livez$,/healthz$,/readyz$,/metrics$) -INFO: 13:05:14 - uvicorn.error - Started server process [1] -INFO: 13:05:14 - uvicorn.error - Waiting for application startup. -INFO: 13:05:18 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:05:18 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] -INFO: 13:05:18 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:05:18 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] -INFO: 13:05:18 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:05:18 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 118619a1d3ed3b201b814dac997d6742 -INFO: 13:05:18 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:05:18 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] -INFO: 13:05:18 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -2026-08-17 13:05:18.510671776 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -INFO: 13:05:18 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:05:18,766 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:05:18,768 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:05:18,830 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:05:18,830 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:05:18,878 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:05:18,878 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx -INFO: 13:05:18 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:05:18 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:05:22 - docling.document_converter - Going to convert document batch... -INFO: 13:05:22 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:05:22 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:05:22 - docling.document_converter - Finished converting document file in 0.43 sec. +INFO: 04:31:43 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:31:43 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:31:43 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:31:43 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:31:43 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/metrics$,/health$,/readyz$,/ready$,/healthz$,/livez$) +INFO: 04:31:43 - uvicorn.error - Started server process [1] +INFO: 04:31:43 - uvicorn.error - Waiting for application startup. +INFO: 04:31:51 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:31:51 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:31:51 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:31:51 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:31:51 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:31:51 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 118619a1d3ed3b201b814dac997d6742 +INFO: 04:31:51 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:31:51 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +INFO: 04:31:51 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +2026-08-24 04:31:51.658302610 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:31:52 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:31:52,075 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:31:52,077 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:31:52,202 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:31:52,203 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:31:52,323 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:31:52,324 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx +INFO: 04:31:52 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:31:52 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:31:56 - docling.document_converter - Going to convert document batch... +INFO: 04:31:56 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:31:56 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:31:56 - docling.document_converter - Finished converting document file in 0.55 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:588: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -WARNING: 13:05:22 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. -INFO: 13:05:22 - docling_jobkit.convert.results - Processed 1 docs in 0.47 seconds. -INFO: 13:05:22 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job d867e3e8-d986-487c-8644-1bdb5b6ec3a9 in 0.47 seconds -INFO: 13:05:22 - uvicorn.access - 172.17.0.1:46654 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:05:22 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:05:22 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:05:22 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:05:22 - docling_serve.app - [TENANT_ID] Task ce340aa2-09b7-4a4f-813f-6067b2a22ac2 created with tenant_id='default' -INFO: 13:05:22 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task ce340aa2-09b7-4a4f-813f-6067b2a22ac2 -INFO: 13:05:22 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:05:22 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 83f6e7119a1499def48d21cd30b03869 -INFO: 13:05:22 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -INFO: 13:05:22 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:05:22,891 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:05:22,891 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:05:22,947 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:05:22,947 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:05:22,996 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:05:22,996 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx -INFO: 13:05:23 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:05:23 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:05:24 - docling.document_converter - Going to convert document batch... -INFO: 13:05:24 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:05:24 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:05:24 - docling.document_converter - Finished converting document file in 0.38 sec. +WARNING: 04:31:56 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:31:56 - docling_jobkit.convert.results - Processed 1 docs in 0.67 seconds. +INFO: 04:31:56 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 312139e4-666e-4776-bcae-f937f46b4874 in 0.67 seconds +INFO: 04:31:58 - uvicorn.access - 172.17.0.1:40360 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:31:58 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:31:58 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:31:58 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:31:58 - docling_serve.app - [TENANT_ID] Task 9ad18297-4c95-491c-8713-0daf142a8fad created with tenant_id='default' +INFO: 04:31:58 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 9ad18297-4c95-491c-8713-0daf142a8fad +INFO: 04:31:58 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:31:58 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 83f6e7119a1499def48d21cd30b03869 +INFO: 04:31:58 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:31:58 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:31:58,692 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:31:58,692 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:31:58,822 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:31:58,823 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:31:58,901 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:31:58,902 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx +INFO: 04:31:59 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:31:59 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:32:01 - docling.document_converter - Going to convert document batch... +INFO: 04:32:01 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:32:01 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:32:01 - docling.document_converter - Finished converting document file in 0.50 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:588: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -INFO: 13:05:24 - docling_jobkit.convert.results - Processed 1 docs in 0.38 seconds. -INFO: 13:05:24 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job ce340aa2-09b7-4a4f-813f-6067b2a22ac2 in 0.38 seconds -INFO: 13:05:24 - uvicorn.access - 172.17.0.1:46654 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:05:24 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:05:24 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:05:24 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:05:24 - docling_serve.app - [TENANT_ID] Task 74f42551-7e32-420d-ae09-a19b5e599616 created with tenant_id='default' -INFO: 13:05:24 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 74f42551-7e32-420d-ae09-a19b5e599616 -INFO: 13:05:24 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:05:24 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 528ff53b5cc66cd5f7f61ef2cd72a8fc -INFO: 13:05:24 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -INFO: 13:05:24 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:05:24,934 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:05:24,935 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx -[INFO] 2026-08-17 13:05:24,996 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:05:24,997 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:05:25,054 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:05:25,055 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx -INFO: 13:05:25 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:05:25 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:05:26 - docling.document_converter - Going to convert document batch... -INFO: 13:05:26 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:05:26 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:05:26 - docling.document_converter - Finished converting document file in 0.38 sec. +INFO: 04:32:01 - docling_jobkit.convert.results - Processed 1 docs in 0.50 seconds. +INFO: 04:32:01 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 9ad18297-4c95-491c-8713-0daf142a8fad in 0.50 seconds +INFO: 04:32:02 - uvicorn.access - 172.17.0.1:40360 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:32:02 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:32:02 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:32:02 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:32:02 - docling_serve.app - [TENANT_ID] Task 6efb0fb6-71b7-47d6-81fa-e5b3e4ea39c0 created with tenant_id='default' +INFO: 04:32:02 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 6efb0fb6-71b7-47d6-81fa-e5b3e4ea39c0 +INFO: 04:32:02 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:32:02 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 528ff53b5cc66cd5f7f61ef2cd72a8fc +INFO: 04:32:02 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:32:02 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:32:02,693 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:32:02,693 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/det/PP-OCRv6_det_small.onnx +[INFO] 2026-08-24 04:32:02,779 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:32:02,779 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:32:02,826 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:32:02,826 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv6/rec/PP-OCRv6_rec_small.onnx +INFO: 04:32:02 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:32:02 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:32:04 - docling.document_converter - Going to convert document batch... +INFO: 04:32:04 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:32:04 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:32:04 - docling.document_converter - Finished converting document file in 0.48 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:588: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -INFO: 13:05:26 - docling_jobkit.convert.results - Processed 1 docs in 0.40 seconds. -INFO: 13:05:26 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 74f42551-7e32-420d-ae09-a19b5e599616 in 0.40 seconds -INFO: 13:05:26 - uvicorn.access - 172.17.0.1:46654 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:05:27 - uvicorn.access - 172.17.0.1:46654 - "GET /v1/clear/converters HTTP/1.1" 200 -INFO: 13:05:27 - uvicorn.access - 172.17.0.1:46654 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 +INFO: 04:32:04 - docling_jobkit.convert.results - Processed 1 docs in 0.53 seconds. +INFO: 04:32:04 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 6efb0fb6-71b7-47d6-81fa-e5b3e4ea39c0 in 0.53 seconds +INFO: 04:32:06 - uvicorn.access - 172.17.0.1:40360 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:32:06 - uvicorn.access - 172.17.0.1:40360 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:32:06 - uvicorn.access - 172.17.0.1:40360 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 ``` @@ -769,136 +952,136 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -INFO: 13:03:29 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:03:29 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] -INFO: 13:03:29 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics -INFO: 13:03:29 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export -INFO: 13:03:29 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/health$,/readyz$,/metrics$,/livez$,/ready$,/healthz$) -INFO: 13:03:29 - uvicorn.error - Started server process [1] -INFO: 13:03:29 - uvicorn.error - Waiting for application startup. -INFO: 13:03:33 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:03:33 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] -INFO: 13:03:33 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:03:33 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] -INFO: 13:03:33 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:03:33 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash c4d6547d10e26a777a530d63ca01a1a1 -INFO: 13:03:33 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:03:33 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] -INFO: 13:03:33 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -2026-08-17 13:03:33.511836816 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -INFO: 13:03:33 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:03:34,016 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:03:34,018 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 13:03:34,116 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:03:34,117 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:03:34,165 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:03:34,165 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 13:03:34 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:03:34 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:03:36 - docling.document_converter - Going to convert document batch... -INFO: 13:03:36 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:03:36 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:03:36 - docling.document_converter - Finished converting document file in 0.42 sec. +INFO: 04:30:15 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:30:15 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'nemotron-ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:30:15 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:30:15 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:30:15 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/livez$,/readyz$,/health$,/healthz$,/metrics$,/ready$) +INFO: 04:30:15 - uvicorn.error - Started server process [1] +INFO: 04:30:15 - uvicorn.error - Waiting for application startup. +INFO: 04:30:18 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:30:18 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:30:18 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:30:18 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:30:18 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:30:18 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash c4d6547d10e26a777a530d63ca01a1a1 +INFO: 04:30:18 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:30:18 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +INFO: 04:30:18 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +2026-08-24 04:30:18.874705432 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:30:19 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:30:19,383 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:30:19,385 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:30:19,489 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:30:19,489 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:30:19,535 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:30:19,535 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:30:19 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:30:19 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:30:21 - docling.document_converter - Going to convert document batch... +INFO: 04:30:21 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:30:21 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:30:22 - docling.document_converter - Finished converting document file in 0.39 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:556: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -WARNING: 13:03:36 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. -INFO: 13:03:36 - docling_jobkit.convert.results - Processed 1 docs in 0.46 seconds. -INFO: 13:03:36 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 48fa076f-eff5-4676-afd7-1c7e840b7872 in 0.46 seconds -INFO: 13:03:37 - uvicorn.access - 172.17.0.1:48118 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:03:37 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:03:37 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:03:37 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:03:37 - docling_serve.app - [TENANT_ID] Task 522dc9bb-2db5-43bf-9c60-a98337d1b2ab created with tenant_id='default' -INFO: 13:03:37 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 522dc9bb-2db5-43bf-9c60-a98337d1b2ab -INFO: 13:03:37 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:03:37 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 4ef9920c0fc80ed745870d6025fc17ac -INFO: 13:03:37 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -INFO: 13:03:37 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:03:37,425 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:03:37,425 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 13:03:37,543 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:03:37,543 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:03:37,591 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:03:37,591 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 13:03:37 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:03:37 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:03:38 - docling.document_converter - Going to convert document batch... -INFO: 13:03:38 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:03:38 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:03:38 - docling.document_converter - Finished converting document file in 0.39 sec. +WARNING: 04:30:22 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:30:22 - docling_jobkit.convert.results - Processed 1 docs in 0.51 seconds. +INFO: 04:30:22 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 51701e21-8620-47e2-8b6a-849b7c241b01 in 0.51 seconds +INFO: 04:30:22 - uvicorn.access - 172.17.0.1:39470 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:30:22 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:30:22 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:30:22 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:30:22 - docling_serve.app - [TENANT_ID] Task 372de451-9a35-4d19-befb-899082bfa691 created with tenant_id='default' +INFO: 04:30:22 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 372de451-9a35-4d19-befb-899082bfa691 +INFO: 04:30:22 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:30:22 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 4ef9920c0fc80ed745870d6025fc17ac +INFO: 04:30:22 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:30:22 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:30:22,724 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:30:22,724 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:30:22,862 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:30:22,862 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:30:22,973 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:30:22,973 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:30:23 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:30:23 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:30:24 - docling.document_converter - Going to convert document batch... +INFO: 04:30:24 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:30:24 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:30:24 - docling.document_converter - Finished converting document file in 0.35 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:556: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -INFO: 13:03:38 - docling_jobkit.convert.results - Processed 1 docs in 0.39 seconds. -INFO: 13:03:38 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 522dc9bb-2db5-43bf-9c60-a98337d1b2ab in 0.39 seconds -INFO: 13:03:39 - uvicorn.access - 172.17.0.1:48118 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:03:39 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:03:39 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:03:39 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:03:39 - docling_serve.app - [TENANT_ID] Task 9d11c878-d211-4186-a542-9b36b63c9352 created with tenant_id='default' -INFO: 13:03:39 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 9d11c878-d211-4186-a542-9b36b63c9352 -INFO: 13:03:39 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:03:39 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 63e8839d1103b03d45382915dba33c89 -INFO: 13:03:39 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. -INFO: 13:03:39 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:03:39,428 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:03:39,428 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 13:03:39,548 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:03:39,548 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:03:39,595 [RapidOCR] base.py:23: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:03:39,596 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 13:03:39 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:03:39 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:03:41 - docling.document_converter - Going to convert document batch... -INFO: 13:03:41 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:03:41 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:03:41 - docling.document_converter - Finished converting document file in 0.40 sec. +INFO: 04:30:24 - docling_jobkit.convert.results - Processed 1 docs in 0.35 seconds. +INFO: 04:30:24 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 372de451-9a35-4d19-befb-899082bfa691 in 0.35 seconds +INFO: 04:30:24 - uvicorn.access - 172.17.0.1:39470 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:30:24 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:30:24 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:30:24 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:30:24 - docling_serve.app - [TENANT_ID] Task 332a5650-bf60-435d-bc4c-eb9dda568980 created with tenant_id='default' +INFO: 04:30:24 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 332a5650-bf60-435d-bc4c-eb9dda568980 +INFO: 04:30:24 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:30:25 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 63e8839d1103b03d45382915dba33c89 +INFO: 04:30:25 - docling.models.stages.ocr.auto_ocr_model - Nemotron cannot be used because it is not installed. +INFO: 04:30:25 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:30:25,086 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:30:25,086 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:30:25,206 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:30:25,207 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:30:25,293 [RapidOCR] base.py:23: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:30:25,293 [RapidOCR] main.py:63: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:30:25 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:30:25 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:30:26 - docling.document_converter - Going to convert document batch... +INFO: 04:30:26 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:30:26 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:30:26 - docling.document_converter - Finished converting document file in 0.45 sec. /opt/app-root/lib64/python3.12/site-packages/docling/datamodel/base_models.py:556: RuntimeWarning: Mean of empty slice np.nanmean( /opt/app-root/lib64/python3.12/site-packages/numpy/lib/_nanfunctions_impl.py:1573: RuntimeWarning: All-NaN slice encountered return _nanquantile_unchecked( -INFO: 13:03:41 - docling_jobkit.convert.results - Processed 1 docs in 0.42 seconds. -INFO: 13:03:41 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 9d11c878-d211-4186-a542-9b36b63c9352 in 0.42 seconds -INFO: 13:03:41 - uvicorn.access - 172.17.0.1:48118 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:03:41 - uvicorn.access - 172.17.0.1:48118 - "GET /v1/clear/converters HTTP/1.1" 200 -INFO: 13:03:41 - uvicorn.access - 172.17.0.1:48118 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 +INFO: 04:30:26 - docling_jobkit.convert.results - Processed 1 docs in 0.51 seconds. +INFO: 04:30:26 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 332a5650-bf60-435d-bc4c-eb9dda568980 in 0.51 seconds +INFO: 04:30:28 - uvicorn.access - 172.17.0.1:39470 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:30:28 - uvicorn.access - 172.17.0.1:39470 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:30:28 - uvicorn.access - 172.17.0.1:39470 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 ``` @@ -936,120 +1119,120 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -INFO: 13:01:52 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:01:52 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] -INFO: 13:01:52 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics -INFO: 13:01:52 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export -INFO: 13:01:52 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/readyz$,/health$,/livez$,/ready$,/healthz$,/metrics$) -INFO: 13:01:52 - uvicorn.error - Started server process [1] -INFO: 13:01:52 - uvicorn.error - Waiting for application startup. -INFO: 13:01:56 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:01:56 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] -INFO: 13:01:56 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:01:56 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] -INFO: 13:01:56 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:01:56 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash f446dcc5a1a7c6aff458e30cab474a9e -INFO: 13:01:56 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:01:56 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] -2026-08-17 13:01:56.746332079 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -INFO: 13:01:57 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:01:57,063 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:01:57,065 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 13:01:57,159 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:01:57,159 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:01:57,208 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:01:57,208 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 13:01:57 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:01:57 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:02:00 - docling.document_converter - Going to convert document batch... -INFO: 13:02:00 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:02:00 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:02:00 - docling.document_converter - Finished converting document file in 0.44 sec. -WARNING: 13:02:00 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. -INFO: 13:02:00 - docling_jobkit.convert.results - Processed 1 docs in 0.48 seconds. -INFO: 13:02:00 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 71d83fd7-e8cd-493d-a245-a5ec1b527502 in 0.48 seconds -INFO: 13:02:00 - uvicorn.access - 172.17.0.1:46044 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:02:00 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:02:00 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:02:00 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:02:00 - docling_serve.app - [TENANT_ID] Task c57c129a-ff25-4a8e-b5b7-f9becac7e959 created with tenant_id='default' -INFO: 13:02:00 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task c57c129a-ff25-4a8e-b5b7-f9becac7e959 -INFO: 13:02:00 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:02:01 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash c639c48e78aacd42b6243acf42ab1c3a -INFO: 13:02:01 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:02:01,065 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:02:01,066 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 13:02:01,191 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:02:01,191 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:02:01,242 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:02:01,242 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 13:02:01 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:02:01 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:02:02 - docling.document_converter - Going to convert document batch... -INFO: 13:02:02 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:02:02 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:02:02 - docling.document_converter - Finished converting document file in 0.40 sec. -INFO: 13:02:02 - docling_jobkit.convert.results - Processed 1 docs in 0.40 seconds. -INFO: 13:02:02 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job c57c129a-ff25-4a8e-b5b7-f9becac7e959 in 0.40 seconds -INFO: 13:02:03 - uvicorn.access - 172.17.0.1:46044 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:02:03 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:02:03 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:02:03 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:02:03 - docling_serve.app - [TENANT_ID] Task 3b683012-c377-460e-8dd5-f414b079fb57 created with tenant_id='default' -INFO: 13:02:03 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 3b683012-c377-460e-8dd5-f414b079fb57 -INFO: 13:02:03 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:02:03 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash e50b6140ec7ee9be0d0fea6ba20c9795 -INFO: 13:02:03 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:02:03,084 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:02:03,084 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 13:02:03,181 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:02:03,181 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:02:03,229 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:02:03,229 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 13:02:03 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:02:03 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:02:04 - docling.document_converter - Going to convert document batch... -INFO: 13:02:04 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:02:04 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:02:04 - docling.document_converter - Finished converting document file in 0.37 sec. -INFO: 13:02:04 - docling_jobkit.convert.results - Processed 1 docs in 0.39 seconds. -INFO: 13:02:04 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 3b683012-c377-460e-8dd5-f414b079fb57 in 0.39 seconds -INFO: 13:02:05 - uvicorn.access - 172.17.0.1:46044 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:02:05 - uvicorn.access - 172.17.0.1:46044 - "GET /v1/clear/converters HTTP/1.1" 200 -INFO: 13:02:05 - uvicorn.access - 172.17.0.1:46044 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 +INFO: 04:29:02 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:29:02 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:29:02 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:29:02 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:29:02 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/health$,/healthz$,/livez$,/readyz$,/metrics$,/ready$) +INFO: 04:29:02 - uvicorn.error - Started server process [1] +INFO: 04:29:02 - uvicorn.error - Waiting for application startup. +INFO: 04:29:07 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:29:07 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:29:07 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:29:07 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:29:07 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:29:07 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash f446dcc5a1a7c6aff458e30cab474a9e +INFO: 04:29:07 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:29:07 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +2026-08-24 04:29:07.906223945 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:29:08 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:29:08,387 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:29:08,390 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:29:08,588 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:29:08,589 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:29:08,717 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:29:08,721 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:29:09 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:29:09 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:29:14 - docling.document_converter - Going to convert document batch... +INFO: 04:29:14 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:29:14 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:29:14 - docling.document_converter - Finished converting document file in 1.13 sec. +WARNING: 04:29:14 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:29:14 - docling_jobkit.convert.results - Processed 1 docs in 1.22 seconds. +INFO: 04:29:14 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job b6694695-21cd-49fa-af9e-bcde1fb1ed30 in 1.22 seconds +INFO: 04:29:15 - uvicorn.access - 172.17.0.1:37238 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:29:15 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:29:15 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:29:15 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:29:15 - docling_serve.app - [TENANT_ID] Task fe2ccf2a-a2e9-4392-bf65-134142a5b771 created with tenant_id='default' +INFO: 04:29:15 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task fe2ccf2a-a2e9-4392-bf65-134142a5b771 +INFO: 04:29:15 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:29:16 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash c639c48e78aacd42b6243acf42ab1c3a +INFO: 04:29:16 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:29:16,296 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:29:16,297 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:29:16,431 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:29:16,431 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:29:16,486 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:29:16,487 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:29:16 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:29:16 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:29:17 - docling.document_converter - Going to convert document batch... +INFO: 04:29:17 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:29:17 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:29:18 - docling.document_converter - Finished converting document file in 0.45 sec. +INFO: 04:29:18 - docling_jobkit.convert.results - Processed 1 docs in 0.45 seconds. +INFO: 04:29:18 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job fe2ccf2a-a2e9-4392-bf65-134142a5b771 in 0.45 seconds +INFO: 04:29:19 - uvicorn.access - 172.17.0.1:37238 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:29:19 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:29:19 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:29:19 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:29:19 - docling_serve.app - [TENANT_ID] Task e354c106-d030-48ce-a465-c9317c1c1221 created with tenant_id='default' +INFO: 04:29:19 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task e354c106-d030-48ce-a465-c9317c1c1221 +INFO: 04:29:19 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:29:19 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash e50b6140ec7ee9be0d0fea6ba20c9795 +INFO: 04:29:19 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:29:19,970 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:29:19,971 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:29:20,069 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:29:20,069 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:29:20,119 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:29:20,119 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:29:20 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:29:20 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:29:21 - docling.document_converter - Going to convert document batch... +INFO: 04:29:21 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:29:21 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:29:21 - docling.document_converter - Finished converting document file in 0.44 sec. +INFO: 04:29:21 - docling_jobkit.convert.results - Processed 1 docs in 0.50 seconds. +INFO: 04:29:21 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job e354c106-d030-48ce-a465-c9317c1c1221 in 0.50 seconds +INFO: 04:29:21 - uvicorn.access - 172.17.0.1:37238 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:29:22 - uvicorn.access - 172.17.0.1:37238 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:29:22 - uvicorn.access - 172.17.0.1:37238 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 ``` @@ -1087,120 +1270,120 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -INFO: 13:00:02 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:00:02 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] -INFO: 13:00:02 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics -INFO: 13:00:02 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export -INFO: 13:00:02 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/ready$,/livez$,/health$,/metrics$,/readyz$,/healthz$) -INFO: 13:00:02 - uvicorn.error - Started server process [1] -INFO: 13:00:02 - uvicorn.error - Waiting for application startup. -INFO: 13:00:06 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:00:06 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] -INFO: 13:00:06 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:00:06 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] -INFO: 13:00:06 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:00:06 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 1743cd6dac10356c82ba7dfffff4ed02 -INFO: 13:00:06 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 13:00:06 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] -2026-08-17 13:00:06.662880127 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -INFO: 13:00:06 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:00:06,975 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:00:06,977 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 13:00:07,072 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:00:07,072 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:00:07,119 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:00:07,119 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 13:00:07 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:00:07 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:00:09 - docling.document_converter - Going to convert document batch... -INFO: 13:00:09 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:00:09 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:00:09 - docling.document_converter - Finished converting document file in 0.44 sec. -WARNING: 13:00:09 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. -INFO: 13:00:09 - docling_jobkit.convert.results - Processed 1 docs in 0.48 seconds. -INFO: 13:00:09 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 235f7cdd-86c3-4870-9818-0122fdf00983 in 0.48 seconds -INFO: 13:00:10 - uvicorn.access - 172.17.0.1:55422 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:00:10 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:00:10 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:00:10 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:00:10 - docling_serve.app - [TENANT_ID] Task e496c771-5102-4016-9a30-a145e58559ad created with tenant_id='default' -INFO: 13:00:10 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task e496c771-5102-4016-9a30-a145e58559ad -INFO: 13:00:10 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:00:10 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 447bb8cb2d04f05cb199adb5a03148da -INFO: 13:00:10 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:00:10,401 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:00:10,401 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 13:00:10,495 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:00:10,495 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:00:10,541 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:00:10,541 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 13:00:10 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:00:10 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:00:12 - docling.document_converter - Going to convert document batch... -INFO: 13:00:12 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:00:12 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:00:12 - docling.document_converter - Finished converting document file in 0.37 sec. -INFO: 13:00:12 - docling_jobkit.convert.results - Processed 1 docs in 0.37 seconds. -INFO: 13:00:12 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job e496c771-5102-4016-9a30-a145e58559ad in 0.37 seconds -INFO: 13:00:12 - uvicorn.access - 172.17.0.1:55422 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:00:12 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 13:00:12 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 13:00:12 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 13:00:12 - docling_serve.app - [TENANT_ID] Task ca8fa13e-8746-425e-bde8-c001dd4263b0 created with tenant_id='default' -INFO: 13:00:12 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task ca8fa13e-8746-425e-bde8-c001dd4263b0 -INFO: 13:00:12 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 13:00:12 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 6102b5a2a09ce02f29b31fc27cc3d04f -INFO: 13:00:12 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 13:00:12,423 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:00:12,424 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 13:00:12,525 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:00:12,526 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 13:00:12,576 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 13:00:12,576 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 13:00:12 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 13:00:12 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 13:00:13 - docling.document_converter - Going to convert document batch... -INFO: 13:00:13 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 13:00:13 - docling.pipeline.base_pipeline - Processing document file -INFO: 13:00:13 - docling.document_converter - Finished converting document file in 0.23 sec. -INFO: 13:00:13 - docling_jobkit.convert.results - Processed 1 docs in 0.25 seconds. -INFO: 13:00:13 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job ca8fa13e-8746-425e-bde8-c001dd4263b0 in 0.25 seconds -INFO: 13:00:14 - uvicorn.access - 172.17.0.1:55422 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 13:00:14 - uvicorn.access - 172.17.0.1:55422 - "GET /v1/clear/converters HTTP/1.1" 200 -INFO: 13:00:14 - uvicorn.access - 172.17.0.1:55422 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 +INFO: 04:27:25 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:27:25 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:27:25 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:27:25 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:27:25 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/metrics$,/ready$,/livez$,/readyz$,/health$,/healthz$) +INFO: 04:27:25 - uvicorn.error - Started server process [1] +INFO: 04:27:25 - uvicorn.error - Waiting for application startup. +INFO: 04:27:29 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:27:29 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:27:29 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:27:29 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:27:29 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:27:29 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 1743cd6dac10356c82ba7dfffff4ed02 +INFO: 04:27:29 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:27:29 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +2026-08-24 04:27:29.615945391 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:27:29 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:27:29,951 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:27:29,952 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:27:30,055 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:27:30,055 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:27:30,101 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:27:30,102 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:27:30 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:27:30 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:27:32 - docling.document_converter - Going to convert document batch... +INFO: 04:27:32 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:27:32 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:27:33 - docling.document_converter - Finished converting document file in 0.74 sec. +WARNING: 04:27:33 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:27:33 - docling_jobkit.convert.results - Processed 1 docs in 0.82 seconds. +INFO: 04:27:33 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 65094f32-cc03-43fc-b25b-3c4520d5e083 in 0.82 seconds +INFO: 04:27:33 - uvicorn.access - 172.17.0.1:59710 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:27:33 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:27:33 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:27:33 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:27:33 - docling_serve.app - [TENANT_ID] Task 13f994da-71df-4ffa-8e63-08834f819399 created with tenant_id='default' +INFO: 04:27:33 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 13f994da-71df-4ffa-8e63-08834f819399 +INFO: 04:27:33 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:27:33 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 447bb8cb2d04f05cb199adb5a03148da +INFO: 04:27:33 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:27:33,454 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:27:33,454 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:27:33,556 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:27:33,556 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:27:33,603 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:27:33,603 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:27:33 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:27:33 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:27:34 - docling.document_converter - Going to convert document batch... +INFO: 04:27:34 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:27:34 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:27:34 - docling.document_converter - Finished converting document file in 0.30 sec. +INFO: 04:27:34 - docling_jobkit.convert.results - Processed 1 docs in 0.30 seconds. +INFO: 04:27:34 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 13f994da-71df-4ffa-8e63-08834f819399 in 0.30 seconds +INFO: 04:27:35 - uvicorn.access - 172.17.0.1:59710 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:27:35 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:27:35 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:27:35 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:27:35 - docling_serve.app - [TENANT_ID] Task 202cc3db-78e5-4566-863b-e4f75c2c9bfe created with tenant_id='default' +INFO: 04:27:35 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 202cc3db-78e5-4566-863b-e4f75c2c9bfe +INFO: 04:27:35 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:27:35 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 6102b5a2a09ce02f29b31fc27cc3d04f +INFO: 04:27:35 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:27:35,457 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:27:35,457 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:27:35,565 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:27:35,565 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:27:35,628 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:27:35,629 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:27:35 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:27:35 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:27:36 - docling.document_converter - Going to convert document batch... +INFO: 04:27:36 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:27:36 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:27:36 - docling.document_converter - Finished converting document file in 0.29 sec. +INFO: 04:27:36 - docling_jobkit.convert.results - Processed 1 docs in 0.34 seconds. +INFO: 04:27:36 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 202cc3db-78e5-4566-863b-e4f75c2c9bfe in 0.34 seconds +INFO: 04:27:37 - uvicorn.access - 172.17.0.1:59710 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:27:37 - uvicorn.access - 172.17.0.1:59710 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:27:37 - uvicorn.access - 172.17.0.1:59710 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 ``` @@ -1238,120 +1421,120 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -INFO: 12:58:37 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:58:37 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] -INFO: 12:58:37 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics -INFO: 12:58:37 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export -INFO: 12:58:37 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/health$,/healthz$,/readyz$,/metrics$,/ready$,/livez$) -INFO: 12:58:37 - uvicorn.error - Started server process [1] -INFO: 12:58:37 - uvicorn.error - Waiting for application startup. -INFO: 12:58:40 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:58:41 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] -INFO: 12:58:41 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:58:41 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] -INFO: 12:58:41 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 12:58:41 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 1743cd6dac10356c82ba7dfffff4ed02 -INFO: 12:58:41 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:58:41 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] -2026-08-17 12:58:41.143898183 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -INFO: 12:58:41 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 12:58:41,442 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:58:41,444 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 12:58:41,539 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:58:41,539 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 12:58:41,585 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:58:41,586 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 12:58:41 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 12:58:41 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 12:58:44 - docling.document_converter - Going to convert document batch... -INFO: 12:58:44 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 12:58:44 - docling.pipeline.base_pipeline - Processing document file -INFO: 12:58:45 - docling.document_converter - Finished converting document file in 0.43 sec. -WARNING: 12:58:45 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. -INFO: 12:58:45 - docling_jobkit.convert.results - Processed 1 docs in 0.49 seconds. -INFO: 12:58:45 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 7c9ac428-2dc8-4831-9c68-cab07ef32a3f in 0.49 seconds -INFO: 12:58:45 - uvicorn.access - 172.17.0.1:55122 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 12:58:45 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 12:58:45 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 12:58:45 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 12:58:45 - docling_serve.app - [TENANT_ID] Task 40d5df01-4bcc-4650-b091-0bf8b7c56589 created with tenant_id='default' -INFO: 12:58:45 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 40d5df01-4bcc-4650-b091-0bf8b7c56589 -INFO: 12:58:45 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 12:58:45 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 447bb8cb2d04f05cb199adb5a03148da -INFO: 12:58:45 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 12:58:45,302 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:58:45,303 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 12:58:45,436 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:58:45,436 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 12:58:45,491 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:58:45,492 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 12:58:45 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 12:58:45 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 12:58:47 - docling.document_converter - Going to convert document batch... -INFO: 12:58:47 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 12:58:47 - docling.pipeline.base_pipeline - Processing document file -INFO: 12:58:47 - docling.document_converter - Finished converting document file in 0.28 sec. -INFO: 12:58:47 - docling_jobkit.convert.results - Processed 1 docs in 0.28 seconds. -INFO: 12:58:47 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 40d5df01-4bcc-4650-b091-0bf8b7c56589 in 0.28 seconds -INFO: 12:58:47 - uvicorn.access - 172.17.0.1:55122 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 12:58:47 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 12:58:47 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 12:58:47 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 12:58:47 - docling_serve.app - [TENANT_ID] Task 4a122485-1d1b-4b3d-994e-b6dc26f103a6 created with tenant_id='default' -INFO: 12:58:47 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 4a122485-1d1b-4b3d-994e-b6dc26f103a6 -INFO: 12:58:47 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 12:58:47 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 6102b5a2a09ce02f29b31fc27cc3d04f -INFO: 12:58:47 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 12:58:47,304 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:58:47,305 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 12:58:47,438 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:58:47,438 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 12:58:47,491 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:58:47,492 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 12:58:47 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 12:58:47 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 12:58:48 - docling.document_converter - Going to convert document batch... -INFO: 12:58:48 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 12:58:48 - docling.pipeline.base_pipeline - Processing document file -INFO: 12:58:48 - docling.document_converter - Finished converting document file in 0.40 sec. -INFO: 12:58:48 - docling_jobkit.convert.results - Processed 1 docs in 0.43 seconds. -INFO: 12:58:48 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 4a122485-1d1b-4b3d-994e-b6dc26f103a6 in 0.43 seconds -INFO: 12:58:49 - uvicorn.access - 172.17.0.1:55122 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 12:58:49 - uvicorn.access - 172.17.0.1:55122 - "GET /v1/clear/converters HTTP/1.1" 200 -INFO: 12:58:49 - uvicorn.access - 172.17.0.1:55122 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 +INFO: 04:26:21 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:26:21 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:26:21 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:26:21 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:26:21 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/health$,/ready$,/livez$,/healthz$,/metrics$,/readyz$) +INFO: 04:26:22 - uvicorn.error - Started server process [1] +INFO: 04:26:22 - uvicorn.error - Waiting for application startup. +INFO: 04:26:28 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:26:28 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:26:28 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:26:28 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:26:28 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:26:28 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 1743cd6dac10356c82ba7dfffff4ed02 +INFO: 04:26:28 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:26:28 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +2026-08-24 04:26:28.781986984 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:26:29 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:26:29,235 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:26:29,237 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:26:29,401 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:26:29,401 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:26:29,468 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:26:29,468 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:26:29 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:26:29 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:26:33 - docling.document_converter - Going to convert document batch... +INFO: 04:26:33 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:26:33 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:26:34 - docling.document_converter - Finished converting document file in 0.53 sec. +WARNING: 04:26:34 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:26:34 - docling_jobkit.convert.results - Processed 1 docs in 0.64 seconds. +INFO: 04:26:34 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 5adecfbb-0ba8-4448-a8e8-8c46329960ef in 0.64 seconds +INFO: 04:26:36 - uvicorn.access - 172.17.0.1:38494 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:26:36 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:26:36 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:26:36 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:26:36 - docling_serve.app - [TENANT_ID] Task f68a8350-673d-44a0-9f2b-4108fdcd65d9 created with tenant_id='default' +INFO: 04:26:36 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task f68a8350-673d-44a0-9f2b-4108fdcd65d9 +INFO: 04:26:36 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:26:36 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 447bb8cb2d04f05cb199adb5a03148da +INFO: 04:26:36 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:26:36,304 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:26:36,304 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:26:36,458 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:26:36,459 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:26:36,516 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:26:36,516 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:26:36 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:26:36 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:26:38 - docling.document_converter - Going to convert document batch... +INFO: 04:26:38 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:26:38 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:26:38 - docling.document_converter - Finished converting document file in 0.49 sec. +INFO: 04:26:38 - docling_jobkit.convert.results - Processed 1 docs in 0.49 seconds. +INFO: 04:26:38 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job f68a8350-673d-44a0-9f2b-4108fdcd65d9 in 0.49 seconds +INFO: 04:26:40 - uvicorn.access - 172.17.0.1:38494 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:26:40 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:26:40 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:26:40 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:26:40 - docling_serve.app - [TENANT_ID] Task b25fb668-551b-455a-82e0-1719d35da987 created with tenant_id='default' +INFO: 04:26:40 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task b25fb668-551b-455a-82e0-1719d35da987 +INFO: 04:26:40 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:26:40 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 6102b5a2a09ce02f29b31fc27cc3d04f +INFO: 04:26:40 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:26:40,310 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:26:40,310 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:26:40,456 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:26:40,456 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:26:40,507 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:26:40,508 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:26:40 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:26:40 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:26:42 - docling.document_converter - Going to convert document batch... +INFO: 04:26:42 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:26:42 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:26:42 - docling.document_converter - Finished converting document file in 0.49 sec. +INFO: 04:26:42 - docling_jobkit.convert.results - Processed 1 docs in 0.53 seconds. +INFO: 04:26:42 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job b25fb668-551b-455a-82e0-1719d35da987 in 0.53 seconds +INFO: 04:26:42 - uvicorn.access - 172.17.0.1:38494 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:26:42 - uvicorn.access - 172.17.0.1:38494 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:26:42 - uvicorn.access - 172.17.0.1:38494 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 ``` @@ -1389,120 +1572,120 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -INFO: 12:56:39 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:56:39 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] -INFO: 12:56:39 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics -INFO: 12:56:39 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export -INFO: 12:56:39 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/health$,/livez$,/healthz$,/readyz$,/metrics$,/ready$) -INFO: 12:56:39 - uvicorn.error - Started server process [1] -INFO: 12:56:39 - uvicorn.error - Waiting for application startup. -INFO: 12:56:43 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:56:43 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] -INFO: 12:56:43 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:56:43 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] -INFO: 12:56:43 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 12:56:43 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 1743cd6dac10356c82ba7dfffff4ed02 -INFO: 12:56:43 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:56:43 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] -2026-08-17 12:56:43.288821848 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -INFO: 12:56:43 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 12:56:43,560 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:56:43,561 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 12:56:43,661 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:56:43,661 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 12:56:43,709 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:56:43,709 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 12:56:43 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 12:56:43 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 12:56:46 - docling.document_converter - Going to convert document batch... -INFO: 12:56:46 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 12:56:46 - docling.pipeline.base_pipeline - Processing document file -INFO: 12:56:46 - docling.document_converter - Finished converting document file in 0.40 sec. -WARNING: 12:56:46 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. -INFO: 12:56:46 - docling_jobkit.convert.results - Processed 1 docs in 0.43 seconds. -INFO: 12:56:46 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 41e37d3f-cf4a-4f16-816f-c32f8a882245 in 0.43 seconds -INFO: 12:56:47 - uvicorn.access - 172.17.0.1:57746 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 12:56:47 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 12:56:47 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 12:56:47 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 12:56:47 - docling_serve.app - [TENANT_ID] Task 3fce9aec-c640-4bea-a584-2d4959fa2961 created with tenant_id='default' -INFO: 12:56:47 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 3fce9aec-c640-4bea-a584-2d4959fa2961 -INFO: 12:56:47 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 12:56:47 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash a1cb01c48360ff351eb870dc682aac59 -INFO: 12:56:47 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 12:56:47,171 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:56:47,171 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 12:56:47,266 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:56:47,267 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 12:56:47,316 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:56:47,317 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 12:56:47 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 12:56:47 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 12:56:48 - docling.document_converter - Going to convert document batch... -INFO: 12:56:48 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 12:56:48 - docling.pipeline.base_pipeline - Processing document file -INFO: 12:56:48 - docling.document_converter - Finished converting document file in 0.53 sec. -INFO: 12:56:48 - docling_jobkit.convert.results - Processed 1 docs in 0.53 seconds. -INFO: 12:56:48 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 3fce9aec-c640-4bea-a584-2d4959fa2961 in 0.53 seconds -INFO: 12:56:49 - uvicorn.access - 172.17.0.1:57746 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 12:56:49 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 12:56:49 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 12:56:49 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 12:56:49 - docling_serve.app - [TENANT_ID] Task 42c98e06-99c8-48db-b0a6-215129f4c524 created with tenant_id='default' -INFO: 12:56:49 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 42c98e06-99c8-48db-b0a6-215129f4c524 -INFO: 12:56:49 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 12:56:49 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash fb1751b09de6bdbe35ede48676ca8f82 -INFO: 12:56:49 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 12:56:49,197 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:56:49,197 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 12:56:49,294 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:56:49,294 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 12:56:49,342 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:56:49,342 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 12:56:49 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 12:56:49 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 12:56:50 - docling.document_converter - Going to convert document batch... -INFO: 12:56:50 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 12:56:50 - docling.pipeline.base_pipeline - Processing document file -INFO: 12:56:50 - docling.document_converter - Finished converting document file in 0.36 sec. -INFO: 12:56:50 - docling_jobkit.convert.results - Processed 1 docs in 0.39 seconds. -INFO: 12:56:50 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 42c98e06-99c8-48db-b0a6-215129f4c524 in 0.39 seconds -INFO: 12:56:51 - uvicorn.access - 172.17.0.1:57746 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 12:56:51 - uvicorn.access - 172.17.0.1:57746 - "GET /v1/clear/converters HTTP/1.1" 200 -INFO: 12:56:51 - uvicorn.access - 172.17.0.1:57746 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 +INFO: 04:24:43 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:24:43 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:24:43 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:24:43 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:24:43 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/metrics$,/ready$,/livez$,/readyz$,/healthz$,/health$) +INFO: 04:24:43 - uvicorn.error - Started server process [1] +INFO: 04:24:43 - uvicorn.error - Waiting for application startup. +INFO: 04:24:46 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:24:46 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:24:46 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:24:46 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:24:46 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:24:46 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 1743cd6dac10356c82ba7dfffff4ed02 +INFO: 04:24:46 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:24:46 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +2026-08-24 04:24:47.016849284 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:24:47 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:24:47,307 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:47,308 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:24:47,399 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:47,399 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:24:47,444 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:47,444 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:24:47 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:24:47 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:24:50 - docling.document_converter - Going to convert document batch... +INFO: 04:24:50 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:24:50 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:24:50 - docling.document_converter - Finished converting document file in 0.43 sec. +WARNING: 04:24:50 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:24:50 - docling_jobkit.convert.results - Processed 1 docs in 0.78 seconds. +INFO: 04:24:50 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 6593b3c8-360f-430e-b869-4b088e0871a1 in 0.78 seconds +INFO: 04:24:50 - uvicorn.access - 172.17.0.1:33300 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:24:50 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:24:50 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:24:50 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:24:50 - docling_serve.app - [TENANT_ID] Task 1e79ab45-503d-40a4-b24e-edf70eb551e5 created with tenant_id='default' +INFO: 04:24:50 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 1e79ab45-503d-40a4-b24e-edf70eb551e5 +INFO: 04:24:50 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:24:50 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash a1cb01c48360ff351eb870dc682aac59 +INFO: 04:24:50 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:24:50,769 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:50,769 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:24:50,863 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:50,863 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:24:50,912 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:50,912 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:24:51 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:24:51 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:24:51 - docling.document_converter - Going to convert document batch... +INFO: 04:24:51 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:24:51 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:24:52 - docling.document_converter - Finished converting document file in 0.30 sec. +INFO: 04:24:52 - docling_jobkit.convert.results - Processed 1 docs in 0.31 seconds. +INFO: 04:24:52 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 1e79ab45-503d-40a4-b24e-edf70eb551e5 in 0.31 seconds +INFO: 04:24:52 - uvicorn.access - 172.17.0.1:33300 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:24:52 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:24:52 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:24:52 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:24:52 - docling_serve.app - [TENANT_ID] Task 2a63cf46-2ebf-4503-8a37-e9720986d48a created with tenant_id='default' +INFO: 04:24:52 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 2a63cf46-2ebf-4503-8a37-e9720986d48a +INFO: 04:24:52 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:24:52 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash fb1751b09de6bdbe35ede48676ca8f82 +INFO: 04:24:52 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:24:52,779 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:52,779 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:24:52,879 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:52,879 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:24:52,925 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:52,925 [RapidOCR] main.py:65: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:24:53 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:24:53 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:24:53 - docling.document_converter - Going to convert document batch... +INFO: 04:24:53 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:24:53 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:24:54 - docling.document_converter - Finished converting document file in 0.29 sec. +INFO: 04:24:54 - docling_jobkit.convert.results - Processed 1 docs in 0.36 seconds. +INFO: 04:24:54 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 2a63cf46-2ebf-4503-8a37-e9720986d48a in 0.36 seconds +INFO: 04:24:54 - uvicorn.access - 172.17.0.1:33300 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:24:55 - uvicorn.access - 172.17.0.1:33300 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:24:55 - uvicorn.access - 172.17.0.1:33300 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 ``` @@ -1540,120 +1723,120 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -INFO: 12:55:23 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:55:23 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] -INFO: 12:55:23 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics -INFO: 12:55:23 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export -INFO: 12:55:23 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/health$,/metrics$,/readyz$,/livez$,/ready$,/healthz$) -INFO: 12:55:23 - uvicorn.error - Started server process [1] -INFO: 12:55:23 - uvicorn.error - Waiting for application startup. -INFO: 12:55:27 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:55:27 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] -INFO: 12:55:27 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:55:27 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] -INFO: 12:55:27 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 12:55:27 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash bcb2e23b24bd94a42b5c00d662a7c1cf -INFO: 12:55:27 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' -INFO: 12:55:27 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] -2026-08-17 12:55:27.965703229 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -INFO: 12:55:28 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 12:55:28,360 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:55:28,362 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 12:55:28,497 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:55:28,497 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 12:55:28,567 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:55:28,568 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 12:55:28 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 12:55:28 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 12:55:32 - docling.document_converter - Going to convert document batch... -INFO: 12:55:32 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 12:55:32 - docling.pipeline.base_pipeline - Processing document file -INFO: 12:55:32 - docling.document_converter - Finished converting document file in 0.60 sec. -WARNING: 12:55:32 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. -INFO: 12:55:32 - docling_jobkit.convert.results - Processed 1 docs in 0.67 seconds. -INFO: 12:55:32 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job b8495b58-463c-40fd-bbb9-e2668f14aa3b in 0.67 seconds -INFO: 12:55:34 - uvicorn.access - 172.17.0.1:36990 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 12:55:34 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 12:55:34 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 12:55:34 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 12:55:34 - docling_serve.app - [TENANT_ID] Task a6039d02-6581-438e-a1f9-eb59be25e0aa created with tenant_id='default' -INFO: 12:55:34 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task a6039d02-6581-438e-a1f9-eb59be25e0aa -INFO: 12:55:34 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 12:55:34 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash c835e945376007df084ecda847e91f5b -INFO: 12:55:34 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 12:55:34,267 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:55:34,268 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 12:55:34,396 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:55:34,396 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 12:55:34,452 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:55:34,453 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 12:55:34 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 12:55:34 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 12:55:36 - docling.document_converter - Going to convert document batch... -INFO: 12:55:36 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 12:55:36 - docling.pipeline.base_pipeline - Processing document file -INFO: 12:55:36 - docling.document_converter - Finished converting document file in 0.71 sec. -INFO: 12:55:36 - docling_jobkit.convert.results - Processed 1 docs in 0.71 seconds. -INFO: 12:55:36 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job a6039d02-6581-438e-a1f9-eb59be25e0aa in 0.71 seconds -INFO: 12:55:38 - uvicorn.access - 172.17.0.1:36990 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 12:55:38 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') -INFO: 12:55:38 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' -INFO: 12:55:38 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata -INFO: 12:55:38 - docling_serve.app - [TENANT_ID] Task e8ce8ce9-1026-49b1-8ba0-d43d1353e8b2 created with tenant_id='default' -INFO: 12:55:38 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task e8ce8ce9-1026-49b1-8ba0-d43d1353e8b2 -INFO: 12:55:38 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. -INFO: 12:55:38 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 3eabf304acb4a56e9ef52b02ef69bccb -INFO: 12:55:38 - docling.utils.accelerator_utils - Accelerator device: 'cpu' -[INFO] 2026-08-17 12:55:38,262 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:55:38,262 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 12:55:38,366 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:55:38,366 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 12:55:38,415 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:55:38,416 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx -INFO: 12:55:38 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. -INFO: 12:55:38 - docling.utils.accelerator_utils - Accelerator device: 'cpu' - Loading weights: 0%| | 0/770 [00:00] -INFO: 12:55:39 - docling.document_converter - Going to convert document batch... -INFO: 12:55:39 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a -INFO: 12:55:39 - docling.pipeline.base_pipeline - Processing document file -INFO: 12:55:40 - docling.document_converter - Finished converting document file in 0.56 sec. -INFO: 12:55:40 - docling_jobkit.convert.results - Processed 1 docs in 0.59 seconds. -INFO: 12:55:40 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job e8ce8ce9-1026-49b1-8ba0-d43d1353e8b2 in 0.59 seconds -INFO: 12:55:40 - uvicorn.access - 172.17.0.1:36990 - "POST /v1/convert/source HTTP/1.1" 200 -INFO: 12:55:40 - uvicorn.access - 172.17.0.1:36990 - "GET /v1/clear/converters HTTP/1.1" 200 -INFO: 12:55:40 - uvicorn.access - 172.17.0.1:36990 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 +INFO: 04:23:47 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:23:47 - docling.models.factories - Registered ocr engines: ['auto', 'easyocr', 'kserve_v2_ocr', 'ocrmac', 'rapidocr', 'tesserocr', 'tesseract'] +INFO: 04:23:47 - docling_serve.otel_instrumentation - Setting up OpenTelemetry metrics +INFO: 04:23:47 - docling_serve.otel_instrumentation - Enabling Prometheus metrics export +INFO: 04:23:47 - docling_serve.otel_instrumentation - Instrumenting FastAPI with OpenTelemetry (excluded_urls=/readyz$,/health$,/metrics$,/healthz$,/ready$,/livez$) +INFO: 04:23:47 - uvicorn.error - Started server process [1] +INFO: 04:23:47 - uvicorn.error - Waiting for application startup. +INFO: 04:23:54 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:23:54 - docling.models.factories - Registered table structure engines: ['docling_tableformer', 'docling_tableformer_v2', 'granite_vision_table'] +INFO: 04:23:54 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:23:54 - docling.models.factories - Registered layout engines: ['layout_object_detection', 'docling_layout_default', 'docling_experimental_table_crops_layout'] +INFO: 04:23:54 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:23:54 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash bcb2e23b24bd94a42b5c00d662a7c1cf +INFO: 04:23:54 - docling.models.factories.base_factory - Loading plugin 'docling_defaults' +INFO: 04:23:54 - docling.models.factories - Registered picture descriptions: ['picture_description_vlm_engine', 'vlm', 'api'] +2026-08-24 04:23:54.371085171 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ +INFO: 04:23:54 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:23:54,808 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:23:54,810 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:23:54,971 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:23:54,971 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:23:55,044 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:23:55,044 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:23:55 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:23:55 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:24:00 - docling.document_converter - Going to convert document batch... +INFO: 04:24:00 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:24:00 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:24:00 - docling.document_converter - Finished converting document file in 0.53 sec. +WARNING: 04:24:00 - docling_core.types.doc.document - Parameter `strict_text` has been deprecated and will be ignored. +INFO: 04:24:00 - docling_jobkit.convert.results - Processed 1 docs in 0.98 seconds. +INFO: 04:24:00 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 65210509-bf9b-416a-9e29-ffd992d21053 in 0.98 seconds +INFO: 04:24:02 - uvicorn.access - 172.17.0.1:49692 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:24:02 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:24:02 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:24:02 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:24:02 - docling_serve.app - [TENANT_ID] Task 59f5c2a6-130e-4d4c-95ab-df3e338db81a created with tenant_id='default' +INFO: 04:24:02 - docling_jobkit.orchestrators.local.worker - Worker 1 processing task 59f5c2a6-130e-4d4c-95ab-df3e338db81a +INFO: 04:24:02 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:24:02 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash c835e945376007df084ecda847e91f5b +INFO: 04:24:02 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:24:02,565 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:02,565 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:24:02,702 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:02,702 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:24:02,807 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:02,807 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:24:02 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:24:02 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:24:04 - docling.document_converter - Going to convert document batch... +INFO: 04:24:04 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:24:04 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:24:04 - docling.document_converter - Finished converting document file in 0.48 sec. +INFO: 04:24:04 - docling_jobkit.convert.results - Processed 1 docs in 0.49 seconds. +INFO: 04:24:04 - docling_jobkit.orchestrators.local.worker - Worker 1 completed job 59f5c2a6-130e-4d4c-95ab-df3e338db81a in 0.49 seconds +INFO: 04:24:04 - uvicorn.access - 172.17.0.1:49692 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:24:04 - docling_serve.app - [TENANT_ID] Extracted tenant_id from header: 'default' (header_value: 'None') +INFO: 04:24:04 - docling_serve.app - [TENANT_ID] process_url endpoint received tenant_id='default' +INFO: 04:24:04 - docling_serve.app - [TENANT_ID] Preparing to enqueue with tenant_id='default' in metadata +INFO: 04:24:04 - docling_serve.app - [TENANT_ID] Task 937e1466-5965-4f10-a5de-a61bcfefe76c created with tenant_id='default' +INFO: 04:24:04 - docling_jobkit.orchestrators.local.worker - Worker 0 processing task 937e1466-5965-4f10-a5de-a61bcfefe76c +INFO: 04:24:04 - docling_jobkit.convert.manager - artifacts_path is set to a valid directory. No model weights will be downloaded at runtime. +INFO: 04:24:04 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash 3eabf304acb4a56e9ef52b02ef69bccb +INFO: 04:24:04 - docling.utils.accelerator_utils - Accelerator device: 'cpu' +[INFO] 2026-08-24 04:24:04,566 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:04,566 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx +[INFO] 2026-08-24 04:24:04,684 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:04,684 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx +[INFO] 2026-08-24 04:24:04,732 [RapidOCR] base.py:22: Using engine_name: onnxruntime +[INFO] 2026-08-24 04:24:04,732 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx +INFO: 04:24:04 - docling.models.stages.ocr.auto_ocr_model - Auto OCR model selected rapidocr with onnxruntime. +INFO: 04:24:04 - docling.utils.accelerator_utils - Accelerator device: 'cpu' + Loading weights: 0%| | 0/770 [00:00] +INFO: 04:24:06 - docling.document_converter - Going to convert document batch... +INFO: 04:24:06 - docling.document_converter - Initializing pipeline for SimplePipeline with options hash 7d306d2d021deac65a97d1a5f925362a +INFO: 04:24:06 - docling.pipeline.base_pipeline - Processing document file +INFO: 04:24:06 - docling.document_converter - Finished converting document file in 0.46 sec. +INFO: 04:24:06 - docling_jobkit.convert.results - Processed 1 docs in 0.52 seconds. +INFO: 04:24:06 - docling_jobkit.orchestrators.local.worker - Worker 0 completed job 937e1466-5965-4f10-a5de-a61bcfefe76c in 0.52 seconds +INFO: 04:24:06 - uvicorn.access - 172.17.0.1:49692 - "POST /v1/convert/source HTTP/1.1" 200 +INFO: 04:24:06 - uvicorn.access - 172.17.0.1:49692 - "GET /v1/clear/converters HTTP/1.1" 200 +INFO: 04:24:06 - uvicorn.access - 172.17.0.1:49692 - "GET /v1/clear/results?older_then=3600 HTTP/1.1" 200 ``` @@ -1691,37 +1874,37 @@ Documentation at http://0.0.0.0:5001/docs Scalar docs at http://0.0.0.0:5001/scalar Logs: -2026-08-17 12:53:28.325152274 [W:onnxruntime:Default, device_discovery.cc:133 GetPciBusId] Skipping pci_bus_id for PCI path at "/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0004:00/MSFT1000:00/5620e0c7-8062-4dce-aeb7-520c7ef76171" because filename "5620e0c7-8062-4dce-aeb7-520c7ef76171" did not match expected pattern of [0-9a-f]+:[0-9a-f]+:[0-9a-f]+[.][0-9a-f]+ -[INFO] 2026-08-17 12:53:28,645 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:53:28,647 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/det/ch_PP-OCRv4_det_mobile.onnx -[INFO] 2026-08-17 12:53:28,747 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:53:28,747 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/cls/ch_ppocr_mobile_v2.0_cls_mobile.onnx -[INFO] 2026-08-17 12:53:28,794 [RapidOCR] base.py:22: Using engine_name: onnxruntime -[INFO] 2026-08-17 12:53:28,795 [RapidOCR] main.py:57: Using /opt/app-root/src/.cache/docling/models/RapidOcr/onnx/PP-OCRv4/rec/ch_PP-OCRv4_rec_mobile.onnx - Loading weights: 0%| | 0/770 [00:00