ONNX Runtime GenAI

Status

Description

Run generative AI models with ONNX Runtime. This API gives you an easy, flexible and performant way of running LLMs on device. It implements the generative AI loop for ONNX models, including pre and post processing, inference with ONNX Runtime, logits processing, search and sampling, KV cache management, and grammar specification for tool calling.

ONNX Runtime GenAI powers Foundry Local, Windows ML, and the Visual Studio Code AI Toolkit.

See documentation at the ONNX Runtime website for more details.

Support matrix	Supported now	Under development	On the roadmap
Model architectures	AMD OLMo ChatGLM DeepSeek ERNIE 4.5 Fara Gemma gpt-oss Granite Granite MoE Hybrid HunYuan Dense V1 InternLM2 Llama Mistral Nemotron Phi (language + vision) Qwen (language + vision) SmolLM3 Whisper	Stable diffusion	Multi-modal models
API	Python C# C/C++ Java ^	Objective-C
O/S	Linux Windows Mac Android		iOS
Architecture	x86 x64 arm64
Hardware Acceleration	CPU CUDA DirectML NvTensorRtRtx (TRT-RTX) OpenVINO QNN WebGPU		AMD GPU
Features	Multi-LoRA Continuous decoding Constrained decoding		Speculative decoding

^ Requires build from source

Installation

See installation instructions or build from source

Sample code for Phi-3 in Python

1. Download the model

   huggingface-cli download microsoft/Phi-3-mini-4k-instruct-onnx --include cpu_and_mobile/cpu-int4-rtn-block-32-acc-level-4/* --local-dir .

2. Install the API

   pip install numpy
   pip install --pre onnxruntime-genai

3. Run the model

   importonnxruntime_genaiasogmodel=og.Model('cpu_and_mobile/cpu-int4-rtn-block-32-acc-level-4')
   tokenizer=og.Tokenizer(model)
   stream=tokenizer.create_stream()
   # Set the max length to something sensible by default,# since otherwise it will be set to the entire context lengthsearch_options= {}
   search_options['max_length'] =2048search_options['batch_size'] =1chat_template='<|user|>\n{input} <|end|>\n<|assistant|>'text=input("Input: ")
   ifnottext:
   print("Error, input cannot be empty")
   exit()
   prompt=f'{chat_template.format(input=text)}'input_tokens=tokenizer.encode(prompt)
   params=og.GeneratorParams(model)
   params.set_search_options(**search_options)
   generator=og.Generator(model, params)
   print("Output: ", end='', flush=True)
   try:
   generator.append_tokens(input_tokens)
   whilenotgenerator.is_done():
   generator.generate_next_token()
   new_token=generator.get_next_tokens()[0]
   print(stream.decode(new_token), end='', flush=True)
   exceptKeyboardInterrupt:
   print(" --control+c pressed, aborting generation--")
   print()
   delgenerator

Choose the correct version of the examples

Due to the evolving nature of this project and ongoing feature additions, examples in the main branch may not always align with the latest stable release. This section outlines how to ensure compatibility between the examples and the corresponding version.

Stable version

Install the package according to the installation instructions. For example, install the Python package.

pip install onnxruntime-genai

Get the version of the package

Linux/Mac:

pip list | grep onnxruntime-genai

Windows:

pip list | findstr "onnxruntime-genai"

Then, check out the version of the examples that corresponds to that release.

# Clone the repo
git clone https://github.com/microsoft/onnxruntime-genai.git &&cd onnxruntime-genai
# Checkout the branch for the version you are using
git checkout v0.11.5
cd examples

Nightly version (main branch)

Checkout the main branch of the repo

git clone https://github.com/microsoft/onnxruntime-genai.git &&cd onnxruntime-genai

Build from source, using these instructions. For example, to build the Python wheel:

python build.py

Navigate to the examples folder in the main branch.

cd examples

To install the nightly Python build:

# Change onnxruntime-genai to the Python package you want to install
pip install --index-url https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/ORT-Nightly/pypi/simple/ onnxruntime-genai

Roadmap

See the Discussions to request new features and up-vote existing requests.

Data/Telemetry

This project may collect usage data and send it to Microsoft to help improve our products and services. See the privacy statement for details.

Contributing

This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.

See DEVELOPMENT.md for how to build, test, and lint the library from source.

When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.

This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.

Linting

This project enables lintrunner for linting. You can install the dependencies and initialize with

pip install -r requirements-lintrunner.txt
lintrunner init

This will install lintrunner on your system and download all the necessary dependencies to run linters locally.

To format local changes:

lintrunner -a

To format all files:

lintrunner -a --all-files

Trademarks

This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.