digitranslab/Megamind-v2-VL-med
Partial fine-tune vs Qwen/Qwen3-VL-8B-Thinking. Partial change: 79% of the weights changed (attention, mlp, norms), in 36 of 36 layers.
Tensors changed: Embeddings 0/4, Norms 89/261, MLP 108/232, Attention 144/252, lm_head 0/1
3 derivatives checked
- mradermacher/Megamind-v2-VL-med-i1-GGUF: Quantization (fewer layers)
- mradermacher/Megamind-v2-VL-med-GGUF: Same weights, converted (fewer layers)
- digitranslab/Megamind-v2-VL-med-gguf: Quantization (fewer layers)
1 other repos hold byte-for-byte identical weight files.