Pin transformers==5.6.2 in install notes

README.md

@@ -20,7 +20,7 @@ Base distilled student; human-level performance. Part of the [PaGE](https://hugg
 ## Self-contained weights
 
 This checkpoint includes the full DINOv3 backbone weights in its `safetensors` files. **No external
-DINOv3 weights are downloaded** — the DINOv3 model *structure* is provided by `transformers>=4.56`
+DINOv3 weights are downloaded** — the DINOv3 model *structure* is provided by `transformers==5.6.2`
 (built-in `dinov3_vit`), and the backbone weights here are derivative weights from full-parameter
 training of DINOv3. The model code (`modeling_page.py`) is loaded automatically from
 [`Octopus1/PaGE`](https://huggingface.co/Octopus1/PaGE) via `auto_map` when you pass
@@ -29,10 +29,10 @@ training of DINOv3. The model code (`modeling_page.py`) is loaded automatically
 ## Installation
 
 ```bash
-pip install torch torchvision timm "transformers>=4.56" safetensors pillow
+pip install torch torchvision timm "transformers==5.6.2" safetensors pillow
 ```
 
-Tested with `transformers` 4.56.0 and 5.6.2.
+Tested with `transformers` 5.6.2.
 
 ## Usage