Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,20 @@
|
|
1 |
-
---
|
2 |
-
license: mit
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
library_name: unity-sentis
|
4 |
+
---
|
5 |
+
|
6 |
+
This is a work-in-progress attempt to use a Phi-3 model with Unity Sentis.
|
7 |
+
|
8 |
+
It's the [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) model converted to .onnx via optimum:
|
9 |
+
|
10 |
+
```
|
11 |
+
from optimum.onnxruntime import ORTModelForCausalLM
|
12 |
+
|
13 |
+
model_id = "microsoft/Phi-3-mini-4k-instruct"
|
14 |
+
model = ORTModelForCausalLM.from_pretrained(model_id, use_cache = False, use_io_binding=False, export=True, trust_remote_code=True, cache_dir=".")
|
15 |
+
model.save_pretrained("phi3_onnx_no_cache/")
|
16 |
+
```
|
17 |
+
|
18 |
+
Then quantized to a Uint8 .sentis model file using Sentis v1.6.0-pre.1
|
19 |
+
|
20 |
+
Usage will be possible with the [com.doji.transformers](https://github.com/julienkay/com.doji.transformers) library, but support for LLMs is not officially released yet.
|