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 supported59 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix][KV Offload] Preserve full specs in scheduler cache projection
- [Bug]: GDN metadata builder hits CUDA device-side assert when adaptive verification trims a mixed prefill+spec batch
- [Bugfix][Kernel] Align packed GDN decode with speculative recurrence
- [Perf][Core] Reserve step budget for waiting requests with cached KV
- [Bug]: collective_rpc timeout does not bound enqueue wait on a full broadcast ring
- [Bugfix][Tool Parser] ernie45: keep content when no complete tool call is found
- [Test][Model Loader] Direct weight reload equals a cold start, bitwise
- [Feature]: Rust frontend: allow fetching multimodal media from private/internal hosts
- [Bugfix][Model Loader] Reload fixes: gpt-oss streamed MXFP4 TP split, w2_bias, FlashInfer bmm scales
- [Bugfix] Fix DeepSeek-V4.1-Flash boot failure on SM120 (RTX PRO 6000 Blackwell): allow 64-page block size for sparse-MLA and indexer backends
- Docs
- Python not yet supported