axiomofmind/Qwen3.8-27B-NVFP4-W4A16-MLP-GGUF
LoRA-style vs Qwen/Qwen3.8-27B. axiomofmind/Qwen3.8-27B-NVFP4-W4A16-MLP-GGUF is a GGUF (T40/Q8_0) of different weights. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj, o_proj, q_proj, v_proj changed, in 16 of 64 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/161, Attention 64/64, Embeddings 0/1, lm_head 0/1