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 supported44 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Distributed] Add MSCCL++ allreduce support for multi-node communication
- [entrypoints/openai] add kv params to streaming api response
- [Perf] Cache get_sm_version_num() in cutlass dispatch
- [Bug]: DeepSeek OCR doesn't work on vllm 0.19
- [Bug]: `response_format` suppresses tool calls when `tool_choice: "auto"` — constrained decoding prevents tool generation
- [Performance]: Eagle3 speculative decoding latency regression in v0.19 vs v0.18
- [Bug]: 投机推理时,异步调度场景,将 -1 作为spec token 赋值给了 input_ids
- [Bug]: Mismatch between batch queue size and max-num-seqs
- [Tracking Issue]: NIXL P/D Disaggregation for Hybrid Models
- fix: allow max_num_batched_tokens > max_model_len with chunked prefill
- Docs
- Python not yet supported