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
- [Security] Defer diffusion prompt scores until output
- [Bugfix][KV Connector] Fail startup when the NIXL listener cannot bind
- [CI] Shard LM eval, MoE kernels and multimodal pooling automatically
- [CI] Shard KV offload, draft decoding, Qwen eval, elastic EP and scheduling automatically
- [Feature]: TTL-based Eviction for KV Cache Blocks
- [Perf][CPU][DiffusionGemma] Run one-step DiffusionGemma reads in a single forward pass
- [ROCm][Perf][Kimi-K3] Mono decode layer for Kimi-K3 on MI355X w/ vanilla decode
- [Bugfix][NIXL] Report failed pulls per request with several KV cache groups
- [Bugfix][NIXL] Re-handshake with a pull peer after a failed read
- [Perf][Mamba] Opt-in FlashInfer SSD prefill scan; per-step ReplaySSM seed
- Docs
- Python not yet supported