mradermacher/GLM-4-T1-54B-GGUF
Quantization vs win10/GLM-4-T1-54B. mradermacher/GLM-4-T1-54B-GGUF is a GGUF quantization (Q8_0) of the same weights: 417 tensors match exactly and 626 within quantization error.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/417, MLP 0/208, Attention 0/416