A new Base only toggle on the Models page hides every finetune, adapter, merge, and quantization, leaving just the original base models.
Need the opposite? The Other тЖТ Model Tree filter does the reverse, listing only one relation type across the Hub, like every adapter or every quantized model.





