AtomicChat/Laguna-S-2.1-GGUF

Quantization vs poolside/Laguna-S-2.1. AtomicChat/Laguna-S-2.1-GGUF is a GGUF quantization (Q8_0) of the same weights: 240 tensors match exactly and 386 within quantization error.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/193, MLP 0/191, Attention 0/240

On the Hub · Downloads on Model Pulse