mradermacher/MT3-Gen4-IF-gemma-2-Riv0.1RAv0.1-9B-GGUF

Same weights, converted vs zelk12/MT3-Gen4-IF-gemma-2-Riv0.1RAv0.1-9B. Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/169, MLP 0/126, Attention 0/168

On the Hub · Downloads on Model Pulse