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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Performance]: int8 W8A8 CUTLASS on Ada (sm_89) is 1.3x to 2x slower per call from M=17, at the dispatch bucket edge
- [Bugfix][Rust] Honor --no-oversample/--disable-shuffle in HF multi-turn
- [Bugfix][KV Offload] Derive the namespace cache dtype from the KV cache specs
- [Performance][MiniMax-M3] Bound sparse decode work and split pages on Blackwell
- [Refactor][Multimodal] Refactor and unify mm prefix range computation
- [Bug]: Per-request EngineDeadError tracebacks accumulate stack frames and flood logs after a worker crash
- [Bugfix][HiSparse][NIXL] Recompute unaligned host-import tails locally
- [Perf][Kernel][Attention] Tile Triton merge_attn_states over tokens
- [Feature]: add NoF (NVMe over Fabrics) replica configuration support to vLLM's MooncakeStoreConnector
- [KV Connector] Support NoF replicas in MooncakeStore
- Docs
- Python not yet supported