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
- fix(security): validate allowed_token_ids against model output vocab size
- fix(security): add token-domain bounds check to penalty bincount kernel
- fix(security): use local variable for grammar in GuidanceBackend.compile.grammar
- [Test] Add fused-MoE-LoRA non-adapter neutrality regression test (refs #49063)
- [ROCm][feature] Add new moe backends supporting int4/int8 weight-only…
- Extend benchmark_moe HF MoE shape parsing beyond Mixtral fallback
- [Bug]: assert req_id in self.requests in Scheduler._update_from_kv_xfer_finished kills the engine on late KV xfer-finished for an already-failed request
- [RFC][SpecDecode] Move MTP completeness validation to the weight-update transaction boundary
- [Bugfix][Granite4Vision] Fix deepstack buffer causing decode slowdown in compiled mode
- [Bug]: response_format=text returns JSON instead of plain text
- Docs
- Python not yet supported