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 supported46 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [ROCm] Support unlimited sequence lengths via multi-pass reduction
- [MoE][Fix] Fix DeepEP HT hardcoded per_act_token_quant=False
- fix(attention): fix high head dim model(Gemma4) support on limited shared memory
- [Frontend] Add /v1/files upload endpoint for multimodal inputs (#38531)
- [Feature]: Built-in request multiplexer to let `vllm serve` use all available GPUs without external proxies
- Entropy-adaptive per-head KV cache quantization: +8% quality over uniform at same compression
- [Bug] vLLM >= 0.18.0 NCCL segfault (cuMemCreate) with TP>1 on RTX 4090 (SM 89)
- [Doc] Update multimodal inputs document about video metadata inputs
- [Bug]:TimeoutError: RPC call to sample_tokens timed out. when pp is on under xpu env
- [Bug]: Regression in vllm 0.19.0 - The page size of the layer is not divisible by the maximum page size
- Docs
- Python not yet supported