deul/Kanana-2-30b-a3b-instruct-2601-GGUF

Quantization vs kakaocorp/kanana-2-30b-a3b-instruct-2601. deul/Kanana-2-30b-a3b-instruct-2601-GGUF is a GGUF quantization (Q8_0) of the same weights: 192 tensors match exactly and 290 within quantization error.

Tensors changed: Embeddings 0/1, Norms 0/145, MLP 0/191, Attention 0/144, lm_head 0/1

On the Hub · Downloads on Model Pulse