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 supported42 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- feat(benchmark): support custom system prompt, request_id tracking, and length filtering controls in bench serve
- [Bug] MXFP8 MoE always falls back to MARLIN on SM_121 (DGX Spark / GB10): TrtLlmFp8ExpertsBase gates on family(100), excluding SM_12x consumer Blackwell
- b12x nvfp4 w4a16 use a16 fix
- [Bug]: DeepSeek reasoning parser can emit reasoning_content and content in the same streaming chunk
- [ROCm][Perf] Enable AR+RMS fusion for GemmaRMSNorm models
- [Bug][Deepseek v4][DBO]: AssertionError: positions is required for C128A metadata build File
- [Bug]: gpt-oss-120b MXFP4 MoE init OOM-killed on unified-memory ARM (DGX Spark / Jetson Thor)
- [Kernel][ROCm] Cover OCP MX MoE emulation in the mxfp4 oracle test
- [Bug]: Qwen3.6-27B-FP8 on GB10: get_output() spin-loop on stuck CUDA stream during prefill burst (run<6, KV<5%)
- MLA: kv_b_proj.weight.dtype AttributeError on quantized ColumnParallelLinear in chunked prefill
- Docs
- Python not yet supported