Update MLX-Gen model card metadata
Browse files
README.md
CHANGED
|
@@ -25,6 +25,10 @@ It uses the mflux/MLX saved-weight layout and MLX quantization tensors. It is no
|
|
| 25 |
|
| 26 |
Original model: [`Qwen/Qwen-Image-Edit`](https://huggingface.co/Qwen/Qwen-Image-Edit).
|
| 27 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
## Quantization
|
| 29 |
|
| 30 |
This is a mixed q4/q8 checkpoint for Qwen Image generation and editing. Fully q4 Qwen checkpoints can lose coherent generative behavior, so MLX-Gen uses a mixed policy:
|
|
|
|
| 25 |
|
| 26 |
Original model: [`Qwen/Qwen-Image-Edit`](https://huggingface.co/Qwen/Qwen-Image-Edit).
|
| 27 |
|
| 28 |
+
## License and Access
|
| 29 |
+
|
| 30 |
+
This quantized derivative follows the Apache 2.0 license of the source model.
|
| 31 |
+
|
| 32 |
## Quantization
|
| 33 |
|
| 34 |
This is a mixed q4/q8 checkpoint for Qwen Image generation and editing. Fully q4 Qwen checkpoints can lose coherent generative behavior, so MLX-Gen uses a mixed policy:
|