meshllm/Muse-Glimmer-30B-UD-Q4_K_XL-layers
Quantization (fewer layers) vs unsloth/Muse-Glimmer-30B-GGUF. meshllm/Muse-Glimmer-30B-UD-Q4_K_XL-layers matches part of unsloth/Muse-Glimmer-30B-GGUF (14 of its 731 tensors; the rest of unsloth/Muse-Glimmer-30B-GGUF has no counterpart). On the shared part: meshllm/Muse-Glimmer-30B-UD-Q4_K_XL-layers is a GGUF quantization (Q4_K/F32) of the same weights: 6 tensors match exactly and 8 within quantization error.
Tensors changed: Norms 0/6, MLP 0/3, Attention 0/5