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 supported29 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- fix(turboquant): use SDPA prefill fallback on pre-Ampere GPUs
- [Kernel] Add prepared-input fast path with MiniMax-M2 top-k/act quant fusion
- attention: pass None for unused args in unified attention TD path
- [Bug]: MiniMax M2.5 TP8EP8 gfx950 with AITER causes memory access fault
- [CI/Perf] Fix serving benchmark JSON validation
- [Bugfix] Expand YAML config when passed as `--config=<path>`
- [Bug]: Accuracy drops ~20% when `--enable-prefix-caching` is used together with MTP speculative decoding (Qwen3.6 35B-A3B)
- [MM][CG] Gemma3 Encoder CUDA Graph
- [Bugfix][ROCm][P/D][MoRIIO] Read-mode KV-release + best_of_n fixes
- [Core][MoE] Add per-activation supports_swiglu_clamp_limit interface
- Docs
- Python not yet supported