samuelcardillo/Qwen3-Coder-Next-Opus-4.6-Reasoning-Distilled

Full fine-tune vs Qwen/Qwen3-Coder-Next. Fine-tune: 93% of the weights changed, across 48 of 48 layers.

Tensors changed: Embeddings 0/1, Norms 147/157, Attention 211/264, MLP 2557/2577, lm_head 1/1

1 derivatives checked

• Momix-44/Huihui-Qwen3-Coder-Next-Opus-4.6-Reasoning-Distilled-abliterated: Abliterated

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

On the Hub · Downloads on Model Pulse