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 supported36 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Humming] Fuse Kimi SiTU activation + per-token FP8 quant on w2 path
- [ROCm][DSV4] Use the AITER tuned GEMM for full-graph decode attention
- [ROCm][DSV4] Opt-in AITER gluon kernel for sparse-MLA decode on gfx950
- [ROCm] Bump AITER to v0.1.19.post2
- [ROCm][AITER] Fix fused QK-norm+RoPE+cache abort by allocating K/V as separate head groups
- [Bugfix] Clamp block table indices in align mode to prevent OOB gather
- [ROCm] Bump AITER to v0.1.19.post2
- [Spec Decode] Fix KV cache group explosion for hybrid models
- [Frontend] Add explicit model info cache preparation
- Revert "[Perf] Narrow DeepSeek V4 eager CUDA graph region" (#51430)
- Docs
- Python not yet supported