mlx-community/Qwen3.5-0.8B-bf16

Full fine-tune vs Qwen/Qwen3.5-0.8B-Base. Fine-tune: 100% of the weights changed, across 24 of 24 layers.

Tensors changed: Attention 126/150, Norms 78/78, Embeddings 1/1, MLP 72/72

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

On the Hub · Downloads on Model Pulse