Image-Text-to-Text
MLX
Safetensors
gemma4_unified
apple-silicon
abliterated
uncensored
crack
vision
audio
omni-modal
gemma4
gemma
google
harmbench
mmlu
jang_4m
conversational
Instructions to use dealignai/Gemma-4-12B-it-JANG_4M-CRACK with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use dealignai/Gemma-4-12B-it-JANG_4M-CRACK with MLX:
# Make sure mlx-vlm is installed # pip install --upgrade mlx-vlm from mlx_vlm import load, generate from mlx_vlm.prompt_utils import apply_chat_template from mlx_vlm.utils import load_config # Load the model model, processor = load("dealignai/Gemma-4-12B-it-JANG_4M-CRACK") config = load_config("dealignai/Gemma-4-12B-it-JANG_4M-CRACK") # Prepare input image = ["http://images.cocodataset.org/val2017/000000039769.jpg"] prompt = "Describe this image." # Apply chat template formatted_prompt = apply_chat_template( processor, config, prompt, num_images=1 ) # Generate output output = generate(model, processor, formatted_prompt, image) print(output) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Pi
How to use dealignai/Gemma-4-12B-it-JANG_4M-CRACK with Pi:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "dealignai/Gemma-4-12B-it-JANG_4M-CRACK"
Configure the model in Pi
# Install Pi: npm install -g @mariozechner/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "mlx-lm": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "dealignai/Gemma-4-12B-it-JANG_4M-CRACK" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use dealignai/Gemma-4-12B-it-JANG_4M-CRACK with Hermes Agent:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "dealignai/Gemma-4-12B-it-JANG_4M-CRACK"
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default dealignai/Gemma-4-12B-it-JANG_4M-CRACK
Run Hermes
hermes
| { | |
| "version": 2, | |
| "weight_format": "jang_affine", | |
| "profile": "JANG_4M", | |
| "source_model": { | |
| "name": "gemma-4-12B-it", | |
| "architecture": "gemma4_unified_text" | |
| }, | |
| "has_vision": true, | |
| "has_audio": true, | |
| "quantization": { | |
| "method": "jang_affine", | |
| "quantization_backend": "mx.quantize", | |
| "mode": "affine", | |
| "group_size": 32, | |
| "tier_bits": { | |
| "attention": 8, | |
| "mlp": 4, | |
| "embed": 16 | |
| }, | |
| "tied_embedding": "fp16_passthrough", | |
| "norm_convention": "gemma4_scale_shift_zero", | |
| "multimodal": "fp16_passthrough_embedders_early_fusion", | |
| "mtp_policy": "none", | |
| "per_module_override_count": 144, | |
| "passthrough_tensor_count": 349 | |
| }, | |
| "runtime": { | |
| "total_weight_bytes": 10135442880, | |
| "total_weight_gb": 9.44, | |
| "attention": "hybrid_swa_full_5to1", | |
| "sliding_window": 1024, | |
| "attention_k_eq_v_on_full_layers": true, | |
| "full_attention_layers": [ | |
| 5, | |
| 11, | |
| 17, | |
| 23, | |
| 29, | |
| 35, | |
| 41, | |
| 47 | |
| ] | |
| }, | |
| "capabilities": { | |
| "reasoning_parser": "gemma4", | |
| "tool_parser": "gemma4", | |
| "think_in_template": false, | |
| "supports_tools": true, | |
| "supports_thinking": true, | |
| "family": "gemma4", | |
| "modality": "vision", | |
| "cache_type": "kv" | |
| } | |
| } |