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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: persistent_topk silently drops top-k candidates when many values share a coarse histogram bin
- [Bugfix] Use FlashInfer auto backend for MXFP4 Linear GEMM instead of hardcoded cute-dsl
- [Bug]: [KV-Transfer] Scheduler crashes due to duplicate `recving` completion events in `MultiConnector` across scheduler steps
- [Bugfix][V1] Fix silent -inf logprobs when logprob_token_ids shares a batch
- [Performance][EPLB] Skip unchanged layers in async transfer cycles
- [Bug]: DeepSeek-V4-Flash-0731 shows severe 1M KV concurrency regression on B300 (27.08x preview vs 7.73x 0731 + DSpark)
- [Bug]: MooncakeStoreConnector process_tokens assertion fails on decode with MTP speculative decoding
- [Test][Core] Cover Mamba-deferred prefix-cache metrics
- [Online quantization] Support activation quant key override
- [Tests][Parser] Add unit tests for tool-call parser metrics
- Docs
- Python not yet supported