InstantX/Qwen-Image-ControlNet-Union

Full fine-tune (fewer layers) vs Qwen/Qwen-Image. InstantX/Qwen-Image-ControlNet-Union matches part of Qwen/Qwen-Image (169 of its 1933 tensors; the rest of Qwen/Qwen-Image has no counterpart). On the shared part: Fine-tune: 91% of the weights changed, across 0 of 0 layers.

Tensors changed: other 23/24, Embeddings 4/4, Attention 76/80, Norms 20/21, MLP 36/40

4 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse