--- title: Gemma-4-12B-it-heretic canonical_url: "https://www.modelscope.ai/models/SC117/Gemma-4-12B-it-heretic" md_url: "https://www.modelscope.ai/models/SC117/Gemma-4-12B-it-heretic.md" repository: SC117/Gemma-4-12B-it-heretic last_updated: 2026-10-04 license: apache-2.0 model_type: - gemma4_unified architectures: - Gemma4UnifiedForConditionalGeneration base_model: - google/gemma-4-12B-it base_model_relation: finetune parameters: 12.0B tensor_type: - F16 library_name: - transformer - safetensors - pytorch frameworks: - pytorch language: - en - zh downloads: 136 stars: 0 tags: - heretic - abliteration - uncensored - gemma4 - safetensors - multimodal - vision - audio --- # Gemma-4-12B-it-heretic > Gemma-4-12B-it-heretic - An open-source model by SC117 on ModelScope. Safetensors Heretic Multimodal Gemma-4-12B-it-heretic English SC117/Gemma-4-12B-it-heretic is a 12.0B-parameter machine learning model on ModelScope. licensed under apache-2.0. derived from google/gemma-4-12B-it. - **Repository**: SC117/Gemma-4-12B-it-heretic - **License**: apache-2.0 - **Parameters**: 12.0B - **Base model**: google/gemma-4-12B-it - **Tags**: heretic, abliteration, uncensored, gemma4, safetensors, multimodal, vision, audio - **Downloads**: 136 - **Stars**: 0 - **Last updated**: 2026-10-04 Source: https://www.modelscope.ai/models/SC117/Gemma-4-12B-it-heretic ---
Safetensors Heretic Multimodal

Gemma-4-12B-it-heretic

English

🧠 Uncensored · Encoder-Free Multimodal · Vision + Audio

Uncensored version of Google Gemma 4 12B IT, processed with Heretic ARA+LoRA abliteration. F32 safetensors format for direct use with Transformers.

Base model: google/gemma-4-12B-it · Parameters: 11.95B · Layers: 48 · Context: 256K · Vocab: 262K

📦 Files
File Size Description
model.safetensors 23 GB F16 model weights
config.json 4.4 KB Model config
tokenizer.json 31 MB Tokenizer
tokenizer_config.json 2.8 KB Tokenizer config
generation_config.json 273 B Generation config
chat_template.jinja 18 KB Chat template
📦 GGUF Quantized Versions

For inference with llama.cpp / Ollama / LM Studio, use the GGUF versions:

• SC117/Gemma-4-12B-it-heretic-GGUF — Q4_K_M / Q6_K / Q8_0 / F16

🚀 Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "SC117/Gemma-4-12B-it-heretic",
    torch_dtype=torch.float32,
    device_map="auto",
)
tokenizer = AutoTokenizer.from_pretrained("SC117/Gemma-4-12B-it-heretic")
📝 Notes

• This is an abliterated (uncensored) model. It may produce content that the original model would refuse. Use responsibly.

• Original model: google/gemma-4-12B-it (Apache 2.0)

Support SC117 on Ko-fi — support is optional and all releases remain freely available.