Add Available classes section to model card

README.md

@@ -68,6 +68,24 @@ Load any DeBERTa variant the same way with `from_weights("kerasformers/<variant>
 | `deberta_v3_base` | [`kerasformers/deberta_v3_base`](https://huggingface.co/kerasformers/deberta_v3_base) | v3 |
 | `deberta_v3_large` | [`kerasformers/deberta_v3_large`](https://huggingface.co/kerasformers/deberta_v3_large) | v3 |
 
+## Available classes
+
+Load any of these from this repo with `from_weights("kerasformers/deberta_v3_base")` (or on the fly via the `hf:` prefix). The pretrained backbone is shared; task heads not stored in this checkpoint start randomly initialized, ready for fine-tuning (or load a `hf:` fine-tune).
+
+| Class | Task |
+|---|---|
+| `DebertaV3Model` | Encoder backbone |
+| `DebertaV3MaskedLM` | Masked language modeling (fill-mask) |
+| `DebertaV3SequenceClassify` | Sequence classification |
+| `DebertaV3TokenClassify` | Token classification (NER / POS) |
+| `DebertaV3QnA` | Extractive question answering |
+| `DebertaV3MultipleChoice` | Multiple choice |
+
+```python
+from kerasformers.models.deberta_v3 import DebertaV3SequenceClassify
+model = DebertaV3SequenceClassify.from_weights("kerasformers/deberta_v3_base")
+```
+
 ## Tips
 
 - Set `KERAS_BACKEND` **before** importing Keras / kerasformers.