← back to catalog · registered 2026-11-12 18:03

metalman3289/gemma-3-12b-it-vl-Gemini-3-Pro-Preview-Heretic-Uncensored-Thinking-mlx-4Bit

metalman3289 Gemma 12B multimodal second-order
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/metalman3289%2Fgemma-3-12b-it-vl-Gemini-3-Pro-Preview-Heretic-Uncensored-Thinking-mlx-4Bit"
Response includes
  • classification m3
  • files 10
  • author_summary 4 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M3
Primary method

Layer-wise ablation

Applied on top of direct removal inherited from the base model.
Confidence
HIGH
Inherited from base model
Why this label 2 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • 'heretic' in model name (Heretic-produced)
  • Heretic uses layer-wise optimization (M3) with underlying direction removal (M1)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
Downloads · 30-day
702
Likes
0
Model age
1w ago
created 2026-11-12

Training datasets

Corpora the author lists in the model card. Datasets tracked in our /datasets catalog carry a category badge linking to the workflow stage. Others open on Hugging Face.

Downloads over time
Now 702 → from 0 ↑ 0%
0257515772 0 on Nov 12 702 on Nov 22 Nov
Nov 12 → Nov 22 · 11 snapshots · spans 10 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Metadata

License
apache-2.0
Languages
en
Tags
transformers safetensors gemma3 image-text-to-text uncensored heretic abliterated unsloth finetune All use cases bfloat16 creative

Related

Total size
6.17 GB
Files
10
Quantizations
1
Registered
2026-11-12 18:03
Last updated on HF
2026-11-12 17:51

Files by quantization

Auxiliary files 10 files 6.20 GB
model-00001-of-00002.safetensors 5.00 GB 16b502d6 download
model-00002-of-00002.safetensors 1.17 GB 933ad3e7 download
tokenizer.json 31.8 MB daab2354 download
model.safetensors.index.json 129 KB 2b90152d download
config.json 3.08 KB 49052c11 download
README.md 1.68 KB b595645a download
.gitattributes 1.53 KB 52373fe2 download
chat_template.jinja 1.50 KB 1117055a download
tokenizer_config.json 715 B 8527b6dc download
generation_config.json 186 B ce59f539 download

README current version from Hugging Face


license: apache-2.0
datasets:

  • TeichAI/gemini-3-pro-preview-high-reasoning-1000x
    language:
  • en
    base_model: DavidAU/gemma-3-12b-it-vl-Gemini-3-Pro-Preview-Heretic-Uncensored-Thinking
    pipeline_tag: image-text-to-text
    library_name: transformers
    tags:
  • uncensored
  • heretic
  • abliterated
  • unsloth
  • finetune
  • All use cases
  • bfloat16
  • creative
  • creative writing
  • fiction writing
  • plot generation
  • sub-plot generation
  • story generation
  • scene continue
  • storytelling
  • fiction story
  • science fiction
  • romance
  • all genres
  • story
  • writing
  • vivid prosing
  • vivid writing
  • fiction
  • mlx
  • mlx-my-repo

metalman3289/gemma-3-12b-it-vl-Gemini-3-Pro-Preview-Heretic-Uncensored-Thinking-mlx-4Bit

The Model metalman3289/gemma-3-12b-it-vl-Gemini-3-Pro-Preview-Heretic-Uncensored-Thinking-mlx-4Bit was converted to MLX format from DavidAU/gemma-3-12b-it-vl-Gemini-3-Pro-Preview-Heretic-Uncensored-Thinking using mlx-lm version 0.31.2.

Use with mlx

pip install mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("metalman3289/gemma-3-12b-it-vl-Gemini-3-Pro-Preview-Heretic-Uncensored-Thinking-mlx-4Bit")

prompt="hello"

if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)

README history 1 version

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-11-12 Upload folder using huggingface_hub 843d5ee 1.7 KB
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration