mradermacher/NQLSG-Qwen2.5-14B-MegaFusion-TIES-4-GGUF

Quantization vs Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-TIES-4. mradermacher/NQLSG-Qwen2.5-14B-MegaFusion-TIES-4-GGUF is a GGUF quantization (Q8_0) of the same weights: 241 tensors match exactly and 338 within quantization error.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/97, MLP 0/144, Attention 0/336

On the Hub · Downloads on Model Pulse