WarmBloodAban/Minimax_H3_LoRAs
Adapter vs MiniMaxAI/MiniMax-H3. WarmBloodAban/Minimax_H3_LoRAs is a LoRA adapter, not full weights: rank 64, on fc1, fc2, out_proj, qkv_proj in 50 layers (208 adapted matrices). It modifies MiniMaxAI/MiniMax-H3 at load time.
1 other repos hold byte-for-byte identical weight files.