litigerking/Hy-MT2-30B-A3B-GGUF

Same weights, converted vs tencent/Hy-MT2-30B-A3B. Same weights stored in another dtype (BF16/F32 → BF16/F32), plus 188 new tensors.

Tensors changed: Norms 0/193, MLP 0/191, Attention 0/192, lm_head 0/1, Embeddings 0/1

On the Hub · Downloads on Model Pulse