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 supported61 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Benchmark][Spec Decode] Add reference-replay oracle benchmark
- [Perf] Avoid quadratic output text copies in the incremental detokenizer
- [Bugfix][Distributed] Register custom AR graph buffers only after a capture that did not raise
- [Bugfix] Guard Conch W4A16 preconditions and pad non-tile-multiple outputs
- [Core] Per-request auxiliary readouts in generation (MRV2)
- [Bug]: Cold-cache KV cache OOM with `--kv-cache-memory-bytes` and `--enable-sleep-mode`
- [Bugfix][Core] Treat a shared-prefix junction as a set of positions under EAGLE
- [Bugfix] Preserve existing files when downloads fail before writing
- [Bug][KV Offload]: Direct-layout offload is treated as parallelism-agnostic, so engines with different TP sizes sharing a secondary tier load corrupted KV
- [Perf][Attention] Fuse DCP correction output layout for reduce-scatter
- Docs
- Python not yet supported