bgg1996/Melinoe-GLM-30B-A3B-Thinking-GGUF
Same weights, converted vs bgg1996/Melinoe-GLM-30B-A3B-Thinking. Same weights stored in another dtype (BF16/F32 → BF16/F32), plus 278 new tensors.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/189, MLP 0/187, Attention 0/188