Run generative AI models with ONNX Runtime.
This library provides the generative AI loop for ONNX models, including inference with ONNX Runtime, logits processing, search and sampling, and KV cache management.
Users can call a high level generate()
method, or run each iteration of the model in a loop.
- Search techniques like greedy/beam search to generate token sequences
- Built in scoring tools like repetition penalties
- Easy custom scoring
Install onnxruntime-genai.
(Temporary) Build and install from source according to the instructions below.
import onnxruntime_genai as og
model=og.Model(f'models/microsoft/phi-2', device_type)
tokenizer = model.CreateTokenizer()
prompt = '''def print_prime(n):
"""
Print all primes between 1 and n
"""'''
tokens = tokenizer.encode(prompt)
params=og.SearchParams(model)
params.max_length = 200
params.input_ids = tokens
output_tokens=model.Generate(params)
text = tokenizer.decode(output_tokens)
print("Output:")
print(text)
- Supported model architectures:
- Phi-2
- Llama
- GPT
- Supported targets:
- CPU
- GPU (CUDA)
- Supported sampling features
- Beam search
- Greedy search
- Top P/Top K
- APIs
- Python
- C/C++
- Support for the Mistral and Whisper model architectures
- C# API
- Support for DirectML
- Automatic model download and cache
- More model architectures
This step requires cmake
to be installed.
-
Clone this repo
git clone https://github.com/microsoft/onnxruntime-genai
-
Install ONNX Runtime
mkdir -p ort/include cd ort/include wget https://raw.githubusercontent.com/microsoft/onnxruntime/v1.17.0/include/onnxruntime/core/session/onnxruntime_c_api.h cd .. mkdir -p ort/lib cd ort/lib wget https://github.com/microsoft/onnxruntime/releases/download/v1.17.0/onnxruntime-linux-x64-gpu-1.17.0.tgz tar xvzf onnxruntime-linux-x64-gpu-1.17.0.tgz cp onnxruntime-linux-x64-gpu-1.17.0/lib/libonnxruntime*.so* .
-
Build onnxruntime-genai
bash build.sh
Or build.bat on Windows
-
Build wheel (temporary)
cd build make PyPackageBuild
-
Install Python wheel
cd wheel pip install *.whl
ONNX models are run from a local folder, via a string supplied to the Model()
method.
To source microsoft/phi-2
optimized for your target, download and run the following script:
wget https://github.com/microsoft/onnxruntime-genai/blob/kvaishnavi/models/models/export.py
Export int4 CPU version
python export.py python models/export.py --m microsoft/phi-2 -p int4 -e cpu -o phi2-int4-cpu.onnx
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.
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 [email protected] with any additional questions or comments.
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.