@vllm_project: The Rust frontend is officially merged into vLLM! As GPUs get faster, the frontend has become a real share of CPU time.…
Summary
The Rust frontend for vLLM has been officially merged, offering a drop-in alternative to the Python API server with up to 5x throughput improvement on preprocess-heavy workloads.
View Cached Full Text
Cached at: 05/27/26, 09:21 AM
The Rust frontend is officially merged into vLLM!
As GPUs get faster, the frontend has become a real share of CPU time. The new Rust frontend is a drop-in alternative to the Python API server — same engine, same ZMQ boundary. Opt in with VLLM_USE_RUST_FRONTEND=1.
Early numbers: on a preprocess-heavy workload, ~837 req/s vs ~162 req/s for default Python — ~5x in a single process.
A few design choices we’re excited about: • Layered crates with clear boundaries • Stream-native pipeline — non-streaming for free • Builds on stable Rust
Huge thanks to @BugenZhao from @inferact for introducing the work at @PyTorch Meetup Singapore.
https://github.com/vllm-project/vllm/pull/40848…
vllm-project/vllm
Source: https://github.com/vllm-project/vllm
Easy, fast, and cheap LLM serving for everyone
| Documentation | Blog | Paper | Twitter/X | User Forum | Developer Slack |
🔥 We have built a vLLM website to help you get started with vLLM. Please visit vllm.ai to learn more. For events, please visit vllm.ai/events to join us.
About
vLLM is a fast and easy-to-use library for LLM inference and serving.
Originally developed in the Sky Computing Lab at UC Berkeley, vLLM has grown into one of the most active open-source AI projects built and maintained by a diverse community of many dozens of academic institutions and companies from over 2000 contributors.
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, chunked prefill, prefix caching
- Fast and flexible model execution with piecewise and full CUDA/HIP graphs
- Quantization: FP8, MXFP8/MXFP4, NVFP4, INT8, INT4, GPTQ/AWQ, GGUF, compressed-tensors, ModelOpt, TorchAO, and more
- Optimized attention kernels including FlashAttention, FlashInfer, TRTLLM-GEN, FlashMLA, and Triton
- Optimized GEMM/MoE kernels for various precisions using CUTLASS, TRTLLM-GEN, CuTeDSL
- Speculative decoding including n-gram, suffix, EAGLE, DFlash
- Automatic kernel generation and graph-level transformations using torch.compile
- Disaggregated prefill, decode, and encode
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, pipeline, data, expert, and context parallelism for distributed inference
- Streaming outputs
- Generation of structured outputs using xgrammar or guidance
- Tool calling and reasoning parsers
- OpenAI-compatible API server, plus Anthropic Messages API and gRPC support
- Efficient multi-LoRA support for dense and MoE layers
- Support for NVIDIA GPUs, AMD GPUs, and x86/ARM/PowerPC CPUs. Additionally, diverse hardware plugins such as Google TPUs, Intel Gaudi, IBM Spyre, Huawei Ascend, Rebellions NPU, Apple Silicon, MetaX GPU, and more.
vLLM seamlessly supports 200+ model architectures on Hugging Face, including:
- Decoder-only LLMs (e.g., Llama, Qwen, Gemma)
- Mixture-of-Expert LLMs (e.g., Mixtral, DeepSeek-V3, Qwen-MoE, GPT-OSS)
- Hybrid attention and state-space models (e.g., Mamba, Qwen3.5)
- Multi-modal models (e.g., LLaVA, Qwen-VL, Pixtral)
- Embedding and retrieval models (e.g., E5-Mistral, GTE, ColBERT)
- Reward and classification models (e.g., Qwen-Math)
Find the full list of supported models here.
Getting Started
Install vLLM with uv (recommended) or pip:
uv pip install vllm
Or build from source for development.
Visit our documentation to learn more.
Contributing
We welcome and value any contributions and collaborations. Please check out Contributing to vLLM for how to get involved.
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}
}
Contact Us
- For technical questions and feature requests, please use GitHub Issues
- For discussing with fellow users, please use the vLLM Forum
- For coordinating contributions and development, please use Slack
- For security disclosures, please use GitHub’s Security Advisories feature
- For collaborations and partnerships, please contact us at [email protected]
Media Kit
- If you wish to use vLLM’s logo, please refer to our media kit repo
Similar Articles
vllm-project/vllm v0.19.1
vLLM v0.19.1 release - a fast and easy-to-use open-source library for LLM inference and serving with state-of-the-art throughput, supporting 200+ model architectures and diverse hardware including NVIDIA/AMD GPUs and CPUs.
Fearless Concurrency on the GPU: Safe GPU inference in Rust, competitive with vLLM/SGLang [R]
cuTile Rust introduces a tile-based programming model that leverages Rust's ownership to guarantee memory safety and data-race freedom for GPU kernels, and the Grout inference engine built on it achieves competitive throughput with vLLM/SGLang for Qwen3 models.
vllm-project/vllm v0.21.0rc1
vLLM v0.21.0rc1 is a pre-release update for the high-performance LLM inference and serving library, featuring optimizations for throughput, quantization, and hardware support.
@binsquares: omg, GPU acceleration on smolvm works way better than I thought. can run llama.cpp inside the smol machine with close t…
User @binsquares reports that GPU acceleration on smolvm achieves nearly 90% of host performance when running llama.cpp via the Vulkan backend.
@QingQ77: Pure Rust LLM inference engine with custom CUDA kernels for each hardware × model × quantization combination, achieving higher inference speed than vLLM and TensorRT-LLM. https://github.com/Avarok-Cybersecurity/a…
Atlas is a pure Rust LLM inference engine that delivers faster inference than vLLM and TensorRT-LLM by customizing CUDA kernels for each hardware × model × quantization combination.