miromind-ai/MiroThinker-1.7
Full fine-tune vs Qwen/Qwen3-235B-A22B-Thinking-2507. Fine-tune: 100% of the weights changed, across 94 of 94 layers.
Tensors changed: Embeddings 1/1, MLP 2245/2245, Norms 377/377, Attention 376/376, lm_head 1/1
2 derivatives checked
- mradermacher/MiroThinker-1.7-i1-GGUF: Quantization
- mradermacher/MiroThinker-1.7-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.