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
- [RFC]: Handle empty responses in synthetic acceptance benchmarks
- [Performance]: Nemotron-3.5-Lightning NVFP4 decode ~16% slower on DGX Spark (GB10/SM121) since v0.29.0
- [RFC]: Octave KV, a native 3-bit KV cache for AMD GPUs
- [Bugfix][Weight Cache] Fingerprint sampled tensor bytes, not headers only
- [Bug]: [Bug]: deepseek_r1 streaming reasoning deltas are not detokenized (byte-level BPE glyphs)
- [Bug]: ~2% output throughput regression on DeepSeek-R1 NVFP4 (DP4 + EP, GB300) from #48247 (AITER custom AG/RS)
- [Metrics] Add delivery counters for the KV event publisher
- [Bugfix][DeepEPv2] Fix ElasticBuffer over-allocated by sp_size× under sequence parallelism
- [MRV2][Spec Decode] Context Aware Sparse LM Head
- [MOE][Quantization] Extend correctness test of `AITER_MXFP4_BF16` MOE backend and temporarily fix correctness issues with Swiglu + no bias code path
- Docs
- Python not yet supported