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 supported36 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [ROCm][CI] Reuse equivalent ROCm CI images
- [wip] add context parallelism for minimax m3 indexer
- transformers backend: per_token_group_quant_8bit not implemented for Float8_e4m3fn (FP8 MoE models)
- transformers backend: tensor reshape error during profile run with GLM MoE architecture
- [Bugfix][DBO] Include ubatch id in attention metadata cache key
- Expose public get_layer_params(config) helper for Gemma4 per-layer attention/FFN params
- [Bug]: `minimax_m3` reasoning parser splits structured-output JSON at a `<mm:think>` the model wrote as content
- [Kernel] Gemma-4 FA4 FP8 Kernel
- Idefics3/SmolVLM: num_patches=0 for non-tiled images mis-sizes pixel_values while the prompt still reserves image_seq_len placeholder tokens
- [V1][Metrics] Preserve prefix-cache stats on zero-output steps
- Docs
- Python not yet supported