DavidAU/gemma-4-31B-it-The-DECKARD-HERETIC-UNCENSORED-Thinking
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 down_proj, gate_proj, k_proj, o_proj, q_proj, up_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 180/261, Attention 230/338
12 derivatives checked
- mradermacher/gemma-4-31B-it-The-DECKARD-HERETIC-UNCENSORED-Thinking-i1-GGUF: Quantization (fewer layers)
- AEON-7/Gemma-4-31B-it-DECKARD-HERETIC-Uncensored-NVFP4: Quantization
- AEON-7/Gemma-4-31B-it-DECKARD-HERETIC-Uncensored-NVFP4-SVDQuant: Quantization
- mradermacher/gemma-4-31B-it-The-DECKARD-HERETIC-UNCENSORED-Thinking-GGUF: Quantization (fewer layers)
- soyaakinohara/gemma-4-31b-it-heretic-3.68bpw-14gb.gguf: Quantization (fewer layers)
- deconfined/gemma-4-31B-it-The-DECKARD-HERETIC-UNCENSORED-Thinking: Quantization
- CountCandy/gemma-4-31B-it-The-DECKARD-HERETIC-UNCENSORED-Thinking-GGUFs: Quantization
- nightmedia/gemma-4-31B-it-The-DECKARD-HERETIC-UNCENSORED-Thinking-qx86-hi-mlx: Quantization
- AEON-7/Gemma-4-E4B-DECKARD-HERETIC-Uncensored-NVFP4: Quantization
- TheCluster/Gemma-4-31B-it-The-DECKARD-HERETIC-UNCENSORED-Thinking-MLX-8bit: Quantization
- Jackxuanxuan/gemma-4-31B-it-The-DECKARD-HERETIC-UNCENSORED-Thinking-i1-GGUF: Quantization (fewer layers)
- kabachuha/Gemma-4-The-Deckards-Brain-31B: LoRA-style
4 other repos hold byte-for-byte identical weight files.