|
---
|
|
license: mit
|
|
pipeline_tag: text-generation
|
|
tags:
|
|
- ONNX
|
|
- ONNXRuntime
|
|
- ONNXRuntimeWeb
|
|
- phi3
|
|
- Transformers.js
|
|
- Transformers
|
|
- nlp
|
|
- conversational
|
|
- custom_code
|
|
inference: false
|
|
---
|
|
|
|
# Phi-3 Mini-4K-Instruct ONNX model for in-browser inference
|
|
|
|
<!-- Provide a quick summary of what the model is/does. -->
|
|
Running Phi3-mini-4K entirely in the browser! Check out this [demo](https://guschmue.github.io/ort-webgpu/chat/index.html).
|
|
|
|
This repository hosts the optimized Web version of ONNX Phi-3-mini-4k-instruct model to accelerate inference in the browser with ONNX Runtime Web.
|
|
|
|
[The Phi-3-Mini-4K-Instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) is a 3.8B parameters, lightweight, state-of-the-art open model trained with the Phi-3 datasets that includes both synthetic data and the filtered publicly available websites data with a focus on high-quality and reasoning dense properties. When assessed against benchmarks testing common sense, language understanding, math, code, long context and logical reasoning, Phi-3 Mini-4K-Instruct showcased a robust and state-of-the-art performance among models with less than 13 billion parameters.
|
|
|
|
## How to run
|
|
|
|
[ONNX Runtime Web](https://onnxruntime.ai/docs/tutorials/web/build-web-app.html) is a JavaScript library to enable web developers to deploy machine learning models directly in web browsers, offering multiple backends leveraging hardware acceleration. WebGPU backend is recommended to run Phi-3-mini efficiently.
|
|
|
|
|
|
Here is an [E2E example](https://github.com/microsoft/onnxruntime-inference-examples/tree/main/js/chat) for running this optimized Phi3-mini-4K for the web, with ONNX Runtime harnessing WebGPU.
|
|
|
|
|
|
**Supported devices and browser with WebGPU**: Chrome 113+ and Edge 113+ for Mac, Windows, ChromeOS, and Chrome 121+ for Android. Pls visit [here](https://github.com/gpuweb/gpuweb/wiki/Implementation-Status#safari-in-progress) for tracking WebGPU support in browsers
|
|
|
|
## Performance Metrics
|
|
Performance vary between GPUs. The more powerful the GPU, the faster the speed. On a NVIDIA GeForce RTX 4090: ~42 tokens/second
|
|
|
|
|
|
## Additional Details
|
|
|
|
To obtain other optimized Phi3-mini-4k ONNX models for server platforms, Windows, Linux, Mac desktops, and mobile, please visit [Phi-3-mini-4k-instruct onnx model](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-onnx). The model differences in the web version compared to other versions:
|
|
|
|
1. the model is fp16 with int4 block quantization for weights
|
|
2. the 'logits' output is fp32
|
|
3. the model uses MHA instead of GQA
|
|
4. onnx and external data file need to stay below 2GB to be cacheable in chromium
|
|
|
|
To optimize a fine-tuned Phi3-mini-4k model to run with ONNX Runtime Web, please follow [this Olive example](https://github.com/microsoft/Olive/tree/main/examples/phi3). [Olive](https://github.com/microsoft/OLive) is an easy-to-use model optimization tool for generating an optimized ONNX model to efficiently run with ONNX Runtime across platforms.
|
|
|
|
|
|
## Model Description
|
|
|
|
- **Developed by:** Microsoft
|
|
- **Model type:** ONNX
|
|
- **Inference Language(s) (NLP):** JavaScript
|
|
- **License:** MIT
|
|
- **Model Description:** This is the web version of the Phi-3 Mini-4K-Instruct model for ONNX Runtime inference.
|
|
|
|
|
|
## Model Card Contact
|
|
guschmue, qining
|
|
|
