meshllm/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL-layers
Quantization (fewer layers) vs unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF. meshllm/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL-layers matches part of unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF (9 of its 927 tensors; the rest of unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF has no counterpart). On the shared part: meshllm/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL-layers is a GGUF quantization (Q8_0/F32) of the same weights: 7 tensors match exactly and 2 within quantization error.
Tensors changed: other 0/7, Norms 0/2