Iris: A Low-Latency Rust/CUDA Runtime for Vision-Language Model Inference.
Iris is built on top of Candle.
Environment: Jetson Orin NX 16 GB (SM87)
| Name | Name | Last commit date | ||
|---|---|---|---|---|
Iris: A Low-Latency Rust/CUDA Runtime for Vision-Language Model Inference.
Iris is built on top of Candle.
Environment: Jetson Orin NX 16 GB (SM87)