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 supported26 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Feature]: aiohttp tracing leaks image urls and cannot be disabled
- [Bug]: V1 engine + MTP + GLM-5.1 (DSA + MoE + MLA) — workers hang under sustained traffic, sample_tokens RPC timeout, EngineDeadError
- [Bug]: --quantization fp8 fails on Qwen3.5 hybrid (qwen3_next gated delta net) with cutlass_scaled_mm Error Internal on GB10 sm_121
- [Bug]: profile_cudagraph_memory() ignores GPU memory clamp on sliced GPUs (HAMi/MIG/MPS) — --gpu-memory-utilization is inert with AutoRound INT4 + fp8_e5m2 KV + FlashInfer + CUDA graphs
- [Bug]: Huggingface Tokenizer "RuntimeError: Already borrowed".
- [Bug]: Running DeepSeek-V4 fails with CUDA error: unspecified launch failure in synchronize_input_prep
- [RFC]: Refactor PassManager infrastructure
- [Usage]: We are using vLLM version 0.19.1. When attempting to run DeepSeek-V4-Flash with a 32k context window across eight RTX 4090 GPUs, we encountered an error indicating that the `transformers` library needed to be updated. We then updated the library using the command `uv pip install --no-cache-dir git+https://github.com/huggingface/transformers.git`, but the error persisted as shown below:
- [Bug]: DeepSeek V4 pro can not run with TP16
- [Tool Parser] Normalize nested arguments in DeepSeek DSML tool calls
- Docs
- Python not yet supported