google/gemma-4-12B-it
Full fine-tune vs google/gemma-4-12B. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 11/12, Norms 193/289, other 44/48, MLP 144/144, Attention 184/184
264 derivatives checked
- unsloth/gemma-4-12b-it-GGUF: Not comparable yet
- HauhauCS/Gemma4-12B-QAT-Uncensored-HauhauCS-Balanced: Full fine-tune
- yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF: Full fine-tune
- lmstudio-community/gemma-4-12B-it-MLX-4bit: Quantization
- yuxinlu1/gemma-4-12B-coder-fable5-composer2.5-v1-GGUF: Full fine-tune
- lmstudio-community/gemma-4-12B-it-MLX-8bit: Quantization
- lmstudio-community/gemma-4-12B-it-MLX-6bit: Quantization
- lmstudio-community/gemma-4-12B-it-MLX-5bit: Quantization
- cyankiwi/gemma-4-12B-it-AWQ-INT4: Quantization
- lmstudio-community/gemma-4-12B-it-GGUF: Quantization
- RedHatAI/gemma-4-12B-it-FP8-Dynamic: Quantization
- EldanRing/Winnow-12B: Not comparable yet
- rahul7star/gemma-gguf: Not comparable yet
- coolthor/gemma-4-12B-it-FP8-dynamic: Quantization
- ggml-org/gemma-4-12B-it-GGUF: Same weights, converted
- unsloth/gemma-4-12b-it: Identical copy
- bartowski/gemma-4-12B-it-GGUF: Same weights, converted
- RedHatAI/gemma-4-12B-it-NVFP4: Quantization
- OBLITERATUS/Gemma-4-12B-OBLITERATED: Partial fine-tune
- google/gemma-4-12B-it-qat-q4_0-unquantized: Full fine-tune
- mattbucci/gemma-4-12B-AWQ: Quantization
- unsloth/gemma-4-12b-it-NVFP4: Quantization
- AEON-7/Gemma-4-12B-it-AEON-Abliterated-K4-BF16: Abliterated
- developerjeremylive/gemma-4-12B-coder-fable5-composer2.5-v1-GGUF-full-etheroi: Full fine-tune
- akhilaaa3/Jev-Omni: LoRA-style
- unsloth/gemma-4-12B-it-qat-q4_0-unquantized: Full fine-tune
- yuxinlu1/gemma-4-12B-it-Claude-4.6-4.8-Opus-GGUF: Full fine-tune
- mlx-community/gemma-4-12B-it-OptiQ-4bit: Quantization
- aleksandard/gemma-4-12B-it-int4-MLPonly-AutoRound: Quantization
- SC117/Gemma-4-12B-it-heretic-GGUF: Abliterated
50 other repos hold byte-for-byte identical weight files.