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
- [Bugfix] Tie lm_head.weight for Nemotron Parse when checkpoint omits it
- [Model][EPLB] Add static expert maps for DeepSeek V4
- [Bug]: `set_splitting_ops_for_v1` early return skips the empty-splitting-ops guard — breakable cudagraph + `cudagraph_mode=PIECEWISE` + spec decode silently rejects every draft
- [Docs] Add troubleshooting for attention backend errors
- [Bug]: MultiConnector: finished_recving lacks per-connector dedup — late/stale load reports crash scheduler assert in P/D + dual-connector setups
- [CI/Build] Extract helper for -static-global-template-stub CUDA flag
- [Bugfix][Kernels] Guard DeepGEMM in mhc_pre_broadcast_tilelang with a torch fallback
- turboquant decode attetnion optimization
- [Bugfix][KV Offload] Fix shared-region ownership and startup cleanup
- [Bugfix][Core] Isolate hidden-state cache from DeepSeek-V4 MLA groups
- Docs
- Python not yet supported