vllm/docs/source/models/extensions/fastsafetensor.md
Reid d6195a748b
[doc] update hyperlink (#16877)
Signed-off-by: reidliu41 <reid201711@gmail.com>
Co-authored-by: reidliu41 <reid201711@gmail.com>
2025-04-19 16:40:38 +00:00

6 lines
418 B
Markdown

Loading Model weights with fastsafetensors
===================================================================
Using fastsafetensors library enables loading model weights to GPU memory by leveraging GPU direct storage. See [their GitHub repository](https://github.com/foundation-model-stack/fastsafetensors) for more details.
For enabling this feature, set the environment variable ``USE_FASTSAFETENSOR`` to ``true``