wimmmm/PrimeIntellect-INTELLECT-3.1-GGUF

Same weights, converted vs PrimeIntellect/INTELLECT-3.1. Same weights stored in another dtype (BF16/F32 → BF16/F32), plus 180 new tensors.

Tensors changed: Embeddings 0/1, Norms 0/93, MLP 0/183, Attention 0/322, lm_head 0/1

On the Hub · Downloads on Model Pulse