Skip to content

nicobasile/vllm

 
 

Repository files navigation

vLLM

Easy, fast, and cheap LLM serving for everyone

| Documentation | Blog | Discussions |


Latest News 🔥

  • [2023/07] Added support for LLaMA-2! You can run and serve 7B/13B/70B LLaMA-2s on vLLM with a single command!
  • [2023/06] Serving vLLM On any Cloud with SkyPilot. Check out a 1-click example to start the vLLM demo, and the blog post for the story behind vLLM development on the clouds.
  • [2023/06] We officially released vLLM! FastChat-vLLM integration has powered LMSYS Vicuna and Chatbot Arena since mid-April. Check out our blog post.

vLLM is a fast and easy-to-use library for LLM inference and serving.

vLLM is fast with:

  • State-of-the-art serving throughput
  • Efficient management of attention key and value memory with PagedAttention
  • Continuous batching of incoming requests
  • Optimized CUDA kernels

vLLM is flexible and easy to use with:

  • Seamless integration with popular HuggingFace models
  • High-throughput serving with various decoding algorithms, including parallel sampling, beam search, and more
  • Tensor parallelism support for distributed inference
  • Streaming outputs
  • OpenAI-compatible API server

vLLM seamlessly supports many Huggingface models, including the following architectures:

  • Baichuan (baichuan-inc/Baichuan-7B, baichuan-inc/Baichuan-13B-Chat, etc.)
  • BLOOM (bigscience/bloom, bigscience/bloomz, etc.)
  • Falcon (tiiuae/falcon-7b, tiiuae/falcon-40b, tiiuae/falcon-rw-7b, etc.)
  • GPT-2 (gpt2, gpt2-xl, etc.)
  • GPT BigCode (bigcode/starcoder, bigcode/gpt_bigcode-santacoder, etc.)
  • GPT-J (EleutherAI/gpt-j-6b, nomic-ai/gpt4all-j, etc.)
  • GPT-NeoX (EleutherAI/gpt-neox-20b, databricks/dolly-v2-12b, stabilityai/stablelm-tuned-alpha-7b, etc.)
  • LLaMA & LLaMA-2 (meta-llama/Llama-2-70b-hf, lmsys/vicuna-13b-v1.3, young-geng/koala, openlm-research/open_llama_13b, etc.)
  • MPT (mosaicml/mpt-7b, mosaicml/mpt-30b, etc.)
  • OPT (facebook/opt-66b, facebook/opt-iml-max-30b, etc.)

Install vLLM with pip or from source:

pip install vllm

Getting Started

Visit our documentation to get started.

Performance

vLLM outperforms HuggingFace Transformers (HF) by up to 24x and Text Generation Inference (TGI) by up to 3.5x, in terms of throughput. For details, check out our blog post.


Serving throughput when each request asks for 1 output completion.


Serving throughput when each request asks for 3 output completions.

Contributing

We welcome and value any contributions and collaborations. Please check out CONTRIBUTING.md for how to get involved.

About

A high-throughput and memory-efficient inference and serving engine for LLMs

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 85.5%
  • Cuda 13.4%
  • Other 1.1%