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 supported60 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Perf][Spec Decode] Vocab-parallel DSpark Markov sampling and vocab-sharded verification over TP
- [Bugfix][ROCm] Fall back to torch.cuda in RocmPlatform when AMDSMI cannot init (#64664)
- [Bugfix][KV Connector][Mooncake] Coalesce in-flight lease heartbeats
- [Bugfix][Spec Decode][XPU] greedy_argmax: treat NaN as -inf on the torch fallback
- [Bugfix][Multimodal] Honour explicit num_frames in Qwen-VL video loaders
- [Model] DeepSeek-V4.1: zero frozen FP8 Engram tables in place in the dummy loader
- [MoE] Release cached allocator blocks before mapping MegaMoE symmetric memory
- [Perf][MoE] Let GateLinear's low-latency router GEMM serve bf16-output routers
- [Perf][MoE] Keep fp32 activation scales on the Triton FP8 experts and re-enable their fused SiLU+quant path
- [ROCm][Perf][DSv4.1] Balance the FP8 paged decode indexer across rows
- Docs
- Python not yet supported