vllm
https://github.com/vllm-project/vllm
Python
A high-throughput and memory-efficient inference and serving engine for LLMs
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported4 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Feature]: Get logits instead of lobprobs for distillation
- [Core] Support all head sizes up to 256 with FlashAttention backend
- [RFC]: QuantizationConfig and QuantizeMethodBase Refactor for Simplifying Kernel Integrations
- [Performance] TTFT regression from v0.5.4 to 0.6.2
- [Bug]: configurably disable prompt echo
- [Performance]: Slowdown compared to Gradio
- [Feature]: Add model context information to chat template
- [Bug]: Server - `aqlm` fails with `--cpu-offload-gb`
- [Bug]: --quantization=awq Using the quantized startup parameters will cause a restart
- [Bug]: assert len(self._async_stopped) == 0
- Docs
- Python not yet supported