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 supported60 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Build] Install infiniband-diags for DeepEPv2 RDMA bandwidth detection
- [Bugfix][KV Offload] Protect the step's reported HITs from promotion eviction
- Fix unbounded series accumulation in vllm:lora_requests_info
- [Bug]: Qwen3.8-27B (GDN hybrid) with PP=3 on sm_86: illegal memory access (Xid 31) with prefix caching, gibberish output without it; regression since v0.28.0 (v0.31.0, v0.32.0, both model runners)
- [Bugfix][MoE] Dynamic per-group BLOCK_M for SM120 DeepGEMM contiguous-grouped MoE
- [Spec Decode] Fix Dspark adaptive verification cost model
- [MyPy] Complete kv-C and enforce KV connector tests
- [LoRA][MoE] Expert LoRA for FlashInfer TRTLLM NVFP4 experts
- [ROCm][Perf] Muse Spark Open Shard residual stream across TP
- [Model] Load q/k/v scales for the quantized Kimi-K2.6
- Docs
- Python not yet supported