OpenMed/Ministral-3B-MedVL
LoRA-style vs mistralai/Ministral-3-3B-Instruct-2512-BF16. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 26 of 26 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/103, MLP 150/150, Attention 200/200, other 0/4
2 derivatives checked
- prithivMLmods/OpenMed-SynthVision-MedVL-AIO-GGUF: Same weights, converted (fewer layers)
- introvoyz041/OpenMed-SynthVision-MedVL-AIO-GGUF: Same weights, converted (fewer layers)