DevQuasar/apple.sage-ft-mixtral-8x7b-GGUF

Same weights, converted vs apple/sage-ft-mixtral-8x7b. Same weights stored in another dtype (BF16 → F16/F32), plus 96 new tensors.

Tensors changed: Embeddings 0/1, other 0/32, Attention 0/128, Norms 0/65, lm_head 0/1

On the Hub · Downloads on Model Pulse