Update README.md
Browse files
README.md
CHANGED
|
@@ -23,6 +23,11 @@ This mistral3 model was trained 2x faster with [Unsloth](https://github.com/unsl
|
|
| 23 |
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
| 24 |
|
| 25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
```python
|
| 27 |
################################################################################
|
| 28 |
# We first load the model for QAT using the mobile CPU friendly int8-int4 scheme
|
|
|
|
| 23 |
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
| 24 |
|
| 25 |
|
| 26 |
+
# Finetune with unsloth and torchao
|
| 27 |
+
|
| 28 |
+
Below we show how to finetune Ministral-3-3B using unsloth in a way that can be deployed with [ExecuTorch](https://github.com/pytorch/executorch).
|
| 29 |
+
The example is based on the notebook [here](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Ministral_3_VL_(3B)_Vision.ipynb#scrollTo=PglJeZZoOWGG).
|
| 30 |
+
|
| 31 |
```python
|
| 32 |
################################################################################
|
| 33 |
# We first load the model for QAT using the mobile CPU friendly int8-int4 scheme
|