AceTheUnfazed/KAT-Coder-V2.5-Dev-APEX-MTP-GGUF
LoRA-style vs mudler/KAT-Coder-V2.5-Dev-APEX-GGUF. AceTheUnfazed/KAT-Coder-V2.5-Dev-APEX-MTP-GGUF is a GGUF (Q3_K/Q4_K) of different weights. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of attn_qkv changed, in 0 of 0 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Norms 0/131, other 0/181, Attention 1/100, MLP 0/320