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
- docs(envs): fix stale VLLM_CPU_KVCACHE_SPACE default comment (follow-up to #60048)
- [Bug]: `--mamba-block-size` has no effect in any valid configuration after the removal of `mamba_cache_mode="all"`
- [Bug]: glm47 streaming parser silently swallows the rest of the stream (including the real tool call) after a stray `<tool_call>` token in prose
- [Feature] Add Winnow read strategy to the Decisions API
- [Bugfix][Parser] Keep a text-spelled reasoning marker as content in non-streaming parse
- [Bugfix][NIXL] Release unreachable prefill peers on failed heartbeat
- [Bugfix] Detect NVFP4 when a compressed-tensors format is a list
- [Tracking] Nemotron Deciffusion 8B (Nemotron Labs Diffusion): native vLLM support, fixes, decision and bidirectional reads
- [Fix][Stacked] Preserve Nemotron diffusion scores and async sampling
- [Bugfix] Drain worker responses before fault-tolerance recovery
- Docs
- Python not yet supported