license: mit
language:
- multilingual
tags:
- nlp
- code
- audio
- automatic-speech-recognition
- speech-summarization
- speech-translation
- visual-question-answering
- phi-4-multimodal
- phi
- phi-4-mini
Phi-4 Multimodal Instruct ONNX models
Introduction
This is an ONNX version of the Phi-4 multimodal model to accelerate inference with ONNX Runtime.
This model is quantized to int4 precision and runs on GPU devices.
To run this model with ONNX Runtime:
Download the model:
git clone https://huggingface.co/microsoft/Phi-4-multimodal-instruct-onnx
Download the script to run the model:
curl https://raw.githubusercontent.com/microsoft/onnxruntime-genai/refs/heads/main/examples/python/phi4-mm.py -o phi4-mm.py
Run the script
python phi4-mm.py -m Phi-4-multimodal-instruct-onnx/gpu/gpu-int4-rtn-block-32 -e cuda
You will be prompted to provide any images, audios, and a prompt.
The performance of the text component is similar to the [Phi-4 mini ONNX models] (https://huggingface.co/microsoft/Phi-4-mini-instruct-onnx/blob/main/README.md)
Model Description
- Developed by: Microsoft
- Model type: ONNX
- License: MIT
- Model Description: This is a conversion of Phi4 mini model for ONNX Runtime inference.
Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for you scenarios. There may be a slight difference in output from the base model with the optimizations applied.
Base Model
Phi-4-multimodal-instruct is a lightweight open multimodal foundation model that leverages the language, vision, and speech research and datasets used for Phi-3.5 and 4.0 models. The model processes text, image, and audio inputs, generating text outputs, and comes with 128K token context length. The model underwent an enhancement process, incorporating both supervised fine-tuning, and direct preference optimization to support precise instruction adherence and safety measures.
See details here