d4rkninja/tanpo-marketing-GGUF
LoRA-style vs LiquidAI/LFM2.5-1.2B-Instruct. d4rkninja/tanpo-marketing-GGUF is a GGUF (F16) of different weights. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj, out_proj, q_proj, v_proj changed, in 6 of 16 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/2, other 0/20, MLP 0/48, Norms 0/44, Attention 24/24