vllm/docs/source/models/extensions/fastsafetensor.md

6 lines
390 B
Markdown
Raw Normal View History

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