cocktailpeanut/stable-audio-3-small-sfx

Partial fine-tune vs stabilityai/stable-audio-3-small-sfx-base. Partial change: 81% of the weights changed (attention, embeddings, norms, other), in 20 of 20 layers.

Tensors changed: Embeddings 95/95, other 105/182, Norms 140/284, Attention 100/124

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

On the Hub · Downloads on Model Pulse