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 supported62 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Attention] FlashMLA sparse: stage fp8 mixed-batch q/out in the workspace
- [Bugfix][Rust Frontend] Clean up failed LoRA loads and handle already-absent unloads
- Refactor RMS norm quant kernel to use vectorized primitives and enable sm_89+ hardware SIMD
- [ROCm][Perf][MLA][DCP] gfx942: run causal DCP verify on AITER single-token asm decode
- [Bug]: auto keeps xgrammar for a combinator with constraint keywords beside it, which xgrammar silently drops
- [Test][Spec Decode] Share reference comparison across EAGLE, MTP and speculators tests
- [Misc] Skip the JIT kernel warmup logs when there is nothing to warm up
- [ROCm][Perf][Kimi-K3] Run 12-head AITER MLA decode natively
- [ROCm][Perf][Kimi-K3] Overlap MoE routing and MLA gate projections
- [Bugfix][Scheduler] Prevent same-step priority eviction
- Docs
- Python not yet supported