mradermacher/Huihui4-8B-A4B-GGUF

Same weights, converted vs huihui-ai/Huihui4-8B-A4B. Same weights stored in another dtype (BF16/F32 → F16/F32), plus 151 new tensors.

Tensors changed: Norms 0/271, Embeddings 0/1, MLP 0/90, other 0/30, Attention 0/115

On the Hub · Downloads on Model Pulse