ji-farthing/Mellum2-12B-A2.5B-Instruct-ik-llama-GGUF

Same weights, converted vs JetBrains/Mellum2-12B-A2.5B-Instruct. Same weights stored in another dtype (BF16 → BF16/F32), plus 84 new tensors.

Tensors changed: Embeddings 0/1, Norms 0/113, MLP 0/28, Attention 0/112, lm_head 0/1

On the Hub · Downloads on Model Pulse