akhilaaa3/Jev-Omni
LoRA-style vs google/gemma-4-12B-it. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/12, Norms 0/289, other 0/48, MLP 144/144, Attention 184/184
4 derivatives checked
- Reza2kn/Jev-Omni-Q4_K_M-GGUF: Quantization
- ngquocvinh/Jev-Omni-GGUF: Quantization
- Ruiruiz30/Jev-Omni-MLX-4bit: Quantization
- ferdinandl007/glim-12b: Identical copy
3 other repos hold byte-for-byte identical weight files.