File size: 833 Bytes
731e725
 
 
 
 
 
82efc80
 
731e725
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
---
license: mit
---

# **Phi-3.5-mini-instruct-onnx-gpu Unofficial version**

<b><span style="text-decoration:underline">Note: This is unoffical version,just for test and dev.</span></b>

This is a Phi-3.5-mini-instruct version of ONNX GPU, based on ONNX Runtime for GenAI [https://github.com/microsoft/onnxruntime-genai](https://github.com/microsoft/onnxruntime-genai). Convert with the following command

## **1. Install the SDK**


```

pip install torch transformers onnx onnxruntime


pip install --pre onnxruntime-genai

```

## **2. Convert GPU ONNX Support**


```bash

python3 -m onnxruntime_genai.models.builder -m microsoft/Phi-3.5-mini-instruct -o ./onnx-gpu -p int4 -e cuda -c ./Phi-3.5-mini-instruct

```



This is a conversion, but no specific optimization has been done. Please look forward to the official version.