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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [V1][LoRA] Version prefix-cache keys for local runtime LoRA reloads
- [Bugfix] Hermes tool parser: parse tool call JSON by object boundary, not literal </tool_call>
- [Bugfix] Fix collect_env.py crash on non-Linux platforms
- [Bug][ROCm/gfx942] GPU memory access fault (worker crash) when sequences cross 2048 tokens — DeepSeek V4 flash arch, sparse_attn_indexer + fp8 KV cache, MI325X TP=4
- [Roadmap] 2026 Q3 vLLM × RL
- [Doc] Clarify that max_num_batched_tokens also shrinks the KV cache pool
- [Bugfix][Tool Parser] Stream tool calls that arrive whole in a single delta in llama3_json, jamba and ernie45 parsers
- [Bugfix][Tool Parser] Mistral v11: keep tool-call boundaries when content precedes multiple tool calls in one delta
- [Bugfix][Frontend] Flatten list message content when echoing in streaming chat completions
- [Bugfix][Tool Parser] DeepSeek V3: don't truncate final streamed tool arguments that don't end with '"}'
- Docs
- Python not yet supported