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
- [Core] Default prefix_cache_retention_interval to a block multiple for hybrid + EAGLE
- [Feature][Frontend] Add opt-in chat message truncation
- [Bugfix][Helm] Add startup probe for slow model initialization
- [Feature][MoE] Collect expert distributions with or without EPLB
- [Performance]: multimodal preprocessing is single-threaded per API server; a wider thread pool does not help (GIL) — process pool for _mm_executor?
- [Bugfix][Model Loader] Keep the weight lock file name within the 255-byte limit
- [Bug]: thinking_token_budget: generation collapses into a repetition loop exactly at budget exhaustion
- [Bug]: Promoted partial prefix-cache entries reference a parent hash removed in the same batch
- [ROCm][Perf][Tracking Issue]: GLM-5.3-Flash
- [RFC] feat(scheduler): implement Shortest Remaining Prefill First (SRPF) policy with starvation aging
- Docs
- Python not yet supported