Easy, fast, and cheap LLM serving for everyone

| Documentation | Blog | Paper | Discord |

---

Setting up vLLM with AQUA

We recommend creating a conda environment with python3>=3.10.12 and gcc=11 to build and test this version of vLLM. Use python3 -m pip install -e . to build and install. conda create --name aquavllm python3=3.10.12 gcc=11 can be used to create the conda environment. Aqua should also be cloned and installed in this environment.

We have introduced the following new flags to enhance scheduling and memory management in vLLM:

AQUA related Command-Line Flags in vLLM:

• --enable-cfs: Enables CFS (Completely Fair Scheduler) for scheduling prompts.
• --enable-aqua-swap: Uses Aqua to create swap space for memory management.
• --be-producer: Configures the GPUs to offer memory for Aqua.
• --producer-req-gb <int>: Specifies the memory availability in GB on each GPU of this GPU.

For details on how to setup producers and consumers with AQUA, please refer to our documentation here - https://aquaml.github.io/

About

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
• Fast model execution with CUDA/HIP graph
• Quantization: GPTQ, AWQ, SqueezeLLM, FP8 KV Cache
• Optimized CUDA kernels

Performance benchmark: We include a performance benchmark that compares the performance of vllm against other LLM serving engines (TensorRT-LLM, text-generation-inference and lmdeploy).

vLLM is flexible and easy to use with:

• Seamless integration with popular Hugging Face models
• High-throughput serving with various decoding algorithms, including parallel sampling, beam search, and more
• Tensor parallelism and pipeline parallelism support for distributed inference
• Streaming outputs
• OpenAI-compatible API server
• Support NVIDIA GPUs, AMD CPUs and GPUs, Intel CPUs and GPUs, PowerPC CPUs
• (Experimental) Prefix caching support
• (Experimental) Multi-lora support

vLLM seamlessly supports most popular open-source models on HuggingFace, including:

• Transformer-like LLMs (e.g., Llama)
• Mixture-of-Expert LLMs (e.g., Mixtral)
• Multi-modal LLMs (e.g., LLaVA)

Find the full list of supported models here.

Getting Started

Install vLLM with pip or from source:

pip install vllm

Visit our documentation to learn more.

• Installation
• Quickstart
• Supported Models

Contributing

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

Sponsors

vLLM is a community project. Our compute resources for development and testing are supported by the following organizations. Thank you for your support!

• a16z
• AMD
• Anyscale
• AWS
• Crusoe Cloud
• Databricks
• DeepInfra
• Dropbox
• Google Cloud
• Lambda Lab
• NVIDIA
• Replicate
• Roblox
• RunPod
• Sequoia Capital
• Trainy
• UC Berkeley
• UC San Diego
• ZhenFund

We also have an official fundraising venue through OpenCollective. We plan to use the fund to support the development, maintenance, and adoption of vLLM.

Citation

If you use vLLM for your research, please cite our paper:

@inproceedings{kwon2023efficient,
  title={Efficient Memory Management for Large Language Model Serving with PagedAttention},
  author={Woosuk Kwon and Zhuohan Li and Siyuan Zhuang and Ying Sheng and Lianmin Zheng and Cody Hao Yu and Joseph E. Gonzalez and Hao Zhang and Ion Stoica},
  booktitle={Proceedings of the ACM SIGOPS 29th Symposium on Operating Systems Principles},
  year={2023}
}