diff --git a/providers/slng/logo.svg b/providers/slng/logo.svg new file mode 100644 index 00000000000..ba8d41fe74a --- /dev/null +++ b/providers/slng/logo.svg @@ -0,0 +1,8 @@ + + + + + + + + diff --git a/providers/slng/models/aura-2.toml b/providers/slng/models/aura-2.toml new file mode 100644 index 00000000000..db5d1ca1a4f --- /dev/null +++ b/providers/slng/models/aura-2.toml @@ -0,0 +1,27 @@ +# Deepgram Aura 2 TTS (Deepgram-hosted via SLNG) +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (AU, EU, US regions, proxied to Deepgram) +# - Original per-minute pricing: $0.0085/min converted to per-token at 195 tokens/min +name = "Aura 2" +description = "Text-to-speech model supporting 7 languages, proxied to Deepgram's infrastructure" +family = "aura" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 43.59 +output = 43.59 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/bulbul-v3.toml b/providers/slng/models/bulbul-v3.toml new file mode 100644 index 00000000000..2977fbcba18 --- /dev/null +++ b/providers/slng/models/bulbul-v3.toml @@ -0,0 +1,27 @@ +# Sarvam AI Bulbul v3 TTS via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (India region) +# - Original per-minute pricing: $0.015/min converted to per-token at 195 tokens/min +name = "Bulbul v3" +description = "Text-to-speech model supporting 11 Indian and global languages" +family = "bulbul" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 76.92 +output = 76.92 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/cartesia-sonic-3-5.toml b/providers/slng/models/cartesia-sonic-3-5.toml new file mode 100644 index 00000000000..2a38a275210 --- /dev/null +++ b/providers/slng/models/cartesia-sonic-3-5.toml @@ -0,0 +1,27 @@ +# Cartesia Sonic 3.5 TTS via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (EU, India, US regions) +# - Original per-minute pricing: $0.02/min converted to per-token at 195 tokens/min +name = "Cartesia Sonic 3.5" +description = "Advanced text-to-speech model supporting 41 languages with enhanced quality" +family = "sonic" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 102.56 +output = 102.56 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/cartesia-sonic-3.toml b/providers/slng/models/cartesia-sonic-3.toml new file mode 100644 index 00000000000..2a376a6c436 --- /dev/null +++ b/providers/slng/models/cartesia-sonic-3.toml @@ -0,0 +1,27 @@ +# Cartesia Sonic 3 TTS via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (EU, India, US regions) +# - Original per-minute pricing: $0.02/min converted to per-token at 195 tokens/min +name = "Cartesia Sonic 3" +description = "Fast text-to-speech model supporting 39 languages with low latency" +family = "sonic" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 102.56 +output = 102.56 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/deepgram-aura-2.toml b/providers/slng/models/deepgram-aura-2.toml new file mode 100644 index 00000000000..da6f1a12a90 --- /dev/null +++ b/providers/slng/models/deepgram-aura-2.toml @@ -0,0 +1,27 @@ +# Deepgram Aura 2 TTS via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (US region, self-hosted) +# - Original per-minute pricing: $0.0085/min converted to per-token at 195 tokens/min +name = "Deepgram Aura 2" +description = "Text-to-speech model for natural voice synthesis in English and Spanish" +family = "aura" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 43.59 +output = 43.59 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/deepgram-nova-3.toml b/providers/slng/models/deepgram-nova-3.toml new file mode 100644 index 00000000000..a62d26f0212 --- /dev/null +++ b/providers/slng/models/deepgram-nova-3.toml @@ -0,0 +1,27 @@ +# Deepgram Nova 3 (Multi) STT via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (India, AU, US regions, self-hosted) +# - Original per-minute pricing: $0.0065/min, capped at $3.00 per million tokens +name = "Deepgram Nova 3 (Multi)" +description = "Multilingual speech-to-text model supporting English, Spanish, and Hindi" +family = "nova" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = false +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 3.00 +output = 3.00 + +[limit] +context = 0 +output = 0 + +[modalities] +input = ["audio"] +output = ["text"] diff --git a/providers/slng/models/fish-tts-s2-1-pro-fish.toml b/providers/slng/models/fish-tts-s2-1-pro-fish.toml new file mode 100644 index 00000000000..cbca6e0f6b8 --- /dev/null +++ b/providers/slng/models/fish-tts-s2-1-pro-fish.toml @@ -0,0 +1,27 @@ +# Fish Audio TTS S2.1 Pro (Fish-hosted via SLNG) +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (US region, proxied to Fish Audio) +# - Fish Audio public pricing: $15.00 per million UTF-8 bytes +name = "Fish TTS S2.1 Pro" +description = "Text-to-speech model supporting 23 languages, proxied to Fish Audio" +family = "fish" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 15.00 +output = 15.00 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/fish-tts-s2-1-pro.toml b/providers/slng/models/fish-tts-s2-1-pro.toml new file mode 100644 index 00000000000..42a33665ad9 --- /dev/null +++ b/providers/slng/models/fish-tts-s2-1-pro.toml @@ -0,0 +1,27 @@ +# Fish Audio TTS S2.1 Pro via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (AU, EU regions, self-hosted) +# - Fish Audio public pricing: $15.00 per million UTF-8 bytes +name = "Fish TTS S2.1 Pro (SLNG)" +description = "Multilingual text-to-speech model supporting 50+ languages, hosted on SLNG infrastructure" +family = "fish" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 15.00 +output = 15.00 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/fish-tts-s2-pro.toml b/providers/slng/models/fish-tts-s2-pro.toml new file mode 100644 index 00000000000..66c0427dc3c --- /dev/null +++ b/providers/slng/models/fish-tts-s2-pro.toml @@ -0,0 +1,27 @@ +# Fish Audio TTS S2 Pro via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (US region, proxied to Fish Audio) +# - Fish Audio public pricing: $15.00 per million UTF-8 bytes +name = "Fish TTS S2 Pro" +description = "Text-to-speech model supporting 23 languages, proxied to Fish Audio" +family = "fish" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 15.00 +output = 15.00 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/gradium-stt.toml b/providers/slng/models/gradium-stt.toml new file mode 100644 index 00000000000..4f65b54d686 --- /dev/null +++ b/providers/slng/models/gradium-stt.toml @@ -0,0 +1,27 @@ +# Gradium STT via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (EU region) +# - Original per-minute pricing: $0.0075/min, capped at $3.00 per million tokens +name = "Gradium STT" +description = "Speech-to-text model supporting German, English, French, Italian, and Portuguese" +family = "gradium" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = false +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 3.00 +output = 3.00 + +[limit] +context = 0 +output = 0 + +[modalities] +input = ["audio"] +output = ["text"] diff --git a/providers/slng/models/gradium-tts.toml b/providers/slng/models/gradium-tts.toml new file mode 100644 index 00000000000..f592dab5dee --- /dev/null +++ b/providers/slng/models/gradium-tts.toml @@ -0,0 +1,27 @@ +# Gradium TTS via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (EU region) +# - Original per-minute pricing: $0.010/min converted to per-token at 195 tokens/min +name = "Gradium TTS" +description = "Text-to-speech model supporting German, English, Spanish, French, Japanese, Korean, and Portuguese" +family = "gradium" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 51.28 +output = 51.28 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/inworld-max-1-5.toml b/providers/slng/models/inworld-max-1-5.toml new file mode 100644 index 00000000000..49e6f0fef66 --- /dev/null +++ b/providers/slng/models/inworld-max-1-5.toml @@ -0,0 +1,27 @@ +# Inworld Max 1.5 TTS via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (EU region, self-hosted) +# - Original per-minute pricing: $0.010/min converted to per-token at 195 tokens/min +name = "Inworld Max 1.5" +description = "Multilingual text-to-speech model supporting 130+ languages" +family = "inworld" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 51.28 +output = 51.28 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/kugel-2.toml b/providers/slng/models/kugel-2.toml new file mode 100644 index 00000000000..e10ca08524a --- /dev/null +++ b/providers/slng/models/kugel-2.toml @@ -0,0 +1,27 @@ +# KugelAudio Kugel 2 TTS via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (EU region) +# - Original per-minute pricing: $0.02/min converted to per-token at 195 tokens/min +name = "Kugel 2" +description = "Text-to-speech model supporting 38 languages" +family = "kugel" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 102.56 +output = 102.56 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/murf-falcon.toml b/providers/slng/models/murf-falcon.toml new file mode 100644 index 00000000000..68e5a94e467 --- /dev/null +++ b/providers/slng/models/murf-falcon.toml @@ -0,0 +1,27 @@ +# Murf Falcon TTS via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (UAE, AU, EU, UK, India, JP, KR, US regions) +# - Original per-minute pricing: $0.0080/min converted to per-token at 195 tokens/min +name = "Murf Falcon" +description = "Text-to-speech model supporting 18 languages across 9 regions" +family = "murf" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 41.03 +output = 41.03 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/nova-3-medical.toml b/providers/slng/models/nova-3-medical.toml new file mode 100644 index 00000000000..8f5e94cdccb --- /dev/null +++ b/providers/slng/models/nova-3-medical.toml @@ -0,0 +1,27 @@ +# Deepgram Nova 3 Medical STT via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (EU, US regions, proxied to Deepgram) +# - Original per-minute pricing: $0.0075/min, capped at $3.00 per million tokens +name = "Nova 3 Medical" +description = "Medical-specialized speech-to-text model for English" +family = "nova" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = false +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 3.00 +output = 3.00 + +[limit] +context = 0 +output = 0 + +[modalities] +input = ["audio"] +output = ["text"] diff --git a/providers/slng/models/nova-3.toml b/providers/slng/models/nova-3.toml new file mode 100644 index 00000000000..774866833f0 --- /dev/null +++ b/providers/slng/models/nova-3.toml @@ -0,0 +1,27 @@ +# Deepgram Nova 3 STT via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (AU, EU, US regions, proxied to Deepgram) +# - Original per-minute pricing: $0.0075/min, capped at $3.00 per million tokens +name = "Nova 3" +description = "Multilingual speech-to-text model supporting 45+ languages" +family = "nova" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = false +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 3.00 +output = 3.00 + +[limit] +context = 0 +output = 0 + +[modalities] +input = ["audio"] +output = ["text"] diff --git a/providers/slng/models/reson8-stt-v1.toml b/providers/slng/models/reson8-stt-v1.toml new file mode 100644 index 00000000000..9e3bb1e1836 --- /dev/null +++ b/providers/slng/models/reson8-stt-v1.toml @@ -0,0 +1,27 @@ +# Reson8 STT v1 via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (EU region) +# - Original per-minute pricing: $0.010/min, capped at $3.00 per million tokens +name = "Reson8 STT v1" +description = "Speech-to-text model for English with specialized terminology support" +family = "reson8" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = false +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 3.00 +output = 3.00 + +[limit] +context = 0 +output = 0 + +[modalities] +input = ["audio"] +output = ["text"] diff --git a/providers/slng/models/sarvam-ai-saaras.toml b/providers/slng/models/sarvam-ai-saaras.toml new file mode 100644 index 00000000000..31936603cd3 --- /dev/null +++ b/providers/slng/models/sarvam-ai-saaras.toml @@ -0,0 +1,27 @@ +# Sarvam AI Saaras STT via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (India region) +# - Original per-minute pricing: $0.0060/min, capped at $3.00 per million tokens +name = "Sarvam AI Saaras" +description = "Speech-to-text model supporting 15 Indian languages" +family = "saaras" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = false +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 3.00 +output = 3.00 + +[limit] +context = 0 +output = 0 + +[modalities] +input = ["audio"] +output = ["text"] diff --git a/providers/slng/models/soniox-speech-ai-rt-v5.toml b/providers/slng/models/soniox-speech-ai-rt-v5.toml new file mode 100644 index 00000000000..a35be4a03c0 --- /dev/null +++ b/providers/slng/models/soniox-speech-ai-rt-v5.toml @@ -0,0 +1,27 @@ +# Soniox Speech AI RT v5 STT via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (EU, Japan, US regions) +# - Original per-minute pricing: $0.0028/min converted at 2000 tokens/min +name = "Speech AI RT v5" +description = "Real-time speech-to-text model supporting 53+ languages" +family = "soniox" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = false +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 1.40 +output = 1.40 + +[limit] +context = 0 +output = 0 + +[modalities] +input = ["audio"] +output = ["text"] diff --git a/providers/slng/models/soniox-tts-rt-v1.toml b/providers/slng/models/soniox-tts-rt-v1.toml new file mode 100644 index 00000000000..f033a931a0b --- /dev/null +++ b/providers/slng/models/soniox-tts-rt-v1.toml @@ -0,0 +1,27 @@ +# Soniox TTS RT v1 via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (EU, JP, US regions) +# - Original per-minute pricing: $0.01/min converted to per-token at 195 tokens/min +name = "Soniox TTS RT v1" +description = "Real-time text-to-speech model supporting 45+ languages" +family = "soniox" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 51.28 +output = 51.28 + +[limit] +context = 32_768 +output = 16_384 + +[modalities] +input = ["text"] +output = ["audio"] diff --git a/providers/slng/models/speechmatics-realtime-v2.toml b/providers/slng/models/speechmatics-realtime-v2.toml new file mode 100644 index 00000000000..d26d2721d0b --- /dev/null +++ b/providers/slng/models/speechmatics-realtime-v2.toml @@ -0,0 +1,27 @@ +# Speechmatics Realtime v2 STT via SLNG +# Sources: +# - https://docs.slng.ai/models — SLNG model catalog (Indonesia, EU regions, self-hosted) +# - Original per-minute pricing: $0.0075/min, capped at $3.00 per million tokens +name = "Speechmatics Realtime v2" +description = "Real-time speech-to-text model for Indonesian language" +family = "speechmatics" +release_date = "2024-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = false +tool_call = false +structured_output = false +open_weights = false + +[cost] +input = 3.00 +output = 3.00 + +[limit] +context = 0 +output = 0 + +[modalities] +input = ["audio"] +output = ["text"] diff --git a/providers/slng/provider.toml b/providers/slng/provider.toml new file mode 100644 index 00000000000..59e69ff17fa --- /dev/null +++ b/providers/slng/provider.toml @@ -0,0 +1,5 @@ +name = "SLNG" +npm = "@ai-sdk/openai-compatible" +api = "https://api.slng.ai/v1" +env = ["SLNG_API_KEY"] +doc = "https://docs.slng.ai/models"