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 supported43 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix] Use block_k for block-wise FP8 activation group_shape
- [Docker] Rename build stages for clarity (base->build-base, build->wheel-build, vllm-base->runtime-base)
- Prewarm full cudagraph capture forward pass
- [codex] Optimize FP8 group 128 quantization for Hopper
- [ROCm][Perf] ROCM_ATTN selected over ROCM_AITER_FA for MHA decode despite VLLM_ROCM_USE_AITER=1 (CDNA)
- [ROCm][Perf] Prefer ROCM_AITER_FA over ROCM_ATTN when AITER MHA is enabled
- [Core] Add scheduler-orchestrated worker transfers for multi-node KV cache tiering
- [KV Connector] Multi-view NIXL connector
- [Bugfix][Spec Decode] Disable dynamic speculative decoding for fixed-K proposers instead of crashing
- [Bug]: Hybrid Mamba PD-disagg under pipeline-parallel (PP>1) emits degenerate output — FullAttention region_group_ids collapse to the Mamba group under HMA
- Docs
- Python not yet supported