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 supported25 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Test] Add nightly MoE eval tests
- fix: check for missing weight files and enable weight integrity for q…
- [BugFix]Fix the bug where query_start_loc values are polluted/corrupted in _dummy_run
- [Bug][ROCm] MI355 + AITER MXFP4 MOE: `Unsupported kernel config for moe heuristic dispatch`
- [kv_offload] Add multi-tier KV cache offloading framework
- Handle capped SWA admission in allocator path
- [Do Not Merge][Attention] Add FLASH_ATTN_MLA_SPARSE backend
- PERF (DeepSeek): Multi-stream overlap of indexer wk+weights_proj with QKV-A for DeepSeek V3.2 NSA
- [Bugfix] Fix uniform_batch for same-sized prefills
- [Bug]: [ROCm] Performance regression in v0.18.2: ROCM_ATTN backend no longer selected by default, falls back to slower TRITON_ATTN
- Docs
- Python not yet supported