mradermacher/Mixtral-8x400M-v0.1-GGUF

Same weights, converted vs Xmaster6y/Mixtral-8x400M-v0.1. Same weights stored in another dtype (F32 → F16/F32), plus 6 new tensors.

Tensors changed: Embeddings 0/1, other 0/2, Attention 0/8, Norms 0/5, lm_head 0/1

On the Hub · Downloads on Model Pulse