mirror of https://github.com/vllm-project/vllm.git
bd5038af07
Signed-off-by: Ernest Wong <chwong719@gmail.com> |
||
|---|---|---|
| .. | ||
| README.md | ||
| faq.md | ||
| metrics.md | ||
| reproducibility.md | ||
| security.md | ||
| troubleshooting.md | ||
| usage_stats.md | ||
| v1_guide.md | ||
README.md
Using vLLM
vLLM supports the following usage patterns:
- Inference and Serving: Run a single instance of a model.
- Deployment: Scale up model instances for production.
- Training: Train or fine-tune a model.