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 supported49 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Feature][Model] Add native DistilBERT sequence classification support
- [Bug][MooncakeConnector] Decode-side bootstrap query has hardcoded 5s httpx default timeout — unconfigurable, no retry, fatal for first PD request in cold-start window
- [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
- [ROCm][Perf][GLM-5.3-Flash]: Tune _rocm_sparse_attn_prefill_ragged_triton for decode
- [Bugfix][CPU] Install spec decode and sampling Triton fallbacks on their dispatchers
- [Feature][MoE] Collect expert distributions with or without EPLB
- [Bugfix] Fix resumable request + async scheduling handoff race
- [Performance]: multimodal preprocessing is single-threaded per API server; a wider thread pool does not help (GIL) — process pool for _mm_executor?
- Docs
- Python not yet supported