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
- [Bugfix][Offloader] Prioritize sparse MoE experts in UVA CPU offloading
- [Bugfix] Fix fp16 precision loss in FunASR position encoding
- [Bug]: NCCL symmetric memory registers comm windows during CUDA graph capture; VLLM_USE_NCCL_SYMM_MEM=1 fails at engine init
- [Core] Apply prefill_schedule_interval outside data parallelism
- [Bug]: LoRA and KV-cache settings reach a compiled torch assertion
- [Bugfix] Fix startup benchmark with empty phases
- [Bugfix][MoE] Fix DP/EP dynamic per-tensor FP8 scales
- [RFC]: fp8 KV cache support for Inkling on SM100-family GPUs
- [Bugfix][Spec Decode] Derive draft KV slots from absolute position on M-RoPE models
- [MoE][Quant] Consider B12X in automatic NVFP4 MoE backend selection
- Docs
- Python not yet supported