DarkKitsune/Qwen3.8-Flash-Next-GSQ-RCO-IQ3_XXS-PLE-50pct
LoRA-style vs ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF. DarkKitsune/Qwen3.8-Flash-Next-GSQ-RCO-IQ3_XXS-PLE-50pct is a GGUF (T42/IQ4_NL) of different weights. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of ffn_down_exps, ffn_gate_exps, ffn_up_exps changed, in 0 of 0 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, other 0/222, Norms 0/184, Attention 0/288, MLP 25/489