PoSTMEDIA/Lux-V2-Pro
LoRA-style vs google/gemma-4-31B-it. 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 60 of 60 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/4, Norms 0/523, other 0/62, MLP 0/261, Attention 230/338
1 derivatives checked
- mradermacher/Lux-V2-Pro-GGUF: Quantization (fewer layers)