1bit-MONSTER/Qwen3.8-Flash-Next-UD-Q4_K_XL-MTP
Full fine-tune vs unsloth/Qwen3.8-Flash-Next-GGUF. 1bit-MONSTER/Qwen3.8-Flash-Next-UD-Q4_K_XL-MTP is a GGUF (Q8_0) of different weights. Fine-tune: 100% of the weights changed, across 0 of 0 layers.
Tensors changed: Attention 9/9, Norms 6/6, MLP 11/11, other 1/1, lm_head 1/1