Get the app
← back to catalog · registered 2026-08-22 13:56

Mori-e/Gemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic-oQ4-fp16

Mori-e Gemma 31B
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/Mori-e%2FGemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic-oQ4-fp16"
Response includes
  • classification m3
  • files 12
  • hub_downloads_all_time 2,278
  • author_summary 3 models
  • readme_text full
Try next
GET/models?author=Mori-e
All 3 models by Mori-e
 GET/models?family=gemma&method=m3
Sibling gemma models · same method
 GET/methods/m3
What "m3" means · taxonomy
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
M1 Direct removal →
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)
How we label → About M3 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
2K
80 last 30d - cooling
Likes
0
Model age
5mo ago
created 2026-05-09
Downloads over time
Now2.3K→from0↑0%
08431.7K2.5K0 on May 62.3K on Oct 12MayJunJulAugSepOct
May 6 → Oct 12 · 63 snapshots · spans 159 days

Metadata

Tags
mlx safetensors gemma4 oq quantized 4-bit region:us

Related

Other by Mori-e
gemma-4-26B-A4B-it-uncensored-oQ5-fp16 160 gemma-4-26B-A4B-it-uncensored-OptiQ 0
Total size
17.7 GB
Files
12
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-05-09 14:18
Source huggingface.co/Mori-e/Gemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic-oQ4-fp16 ↗

Files by quantization

Auxiliary files 12 files 17.7 GB
model-00002-of-00004.safetensors 4.68 GB c02c2ae3 download
model-00003-of-00004.safetensors 4.66 GB 7369ac3a download
model-00001-of-00004.safetensors 4.66 GB 890459df download
model-00004-of-00004.safetensors 3.70 GB 54cb6a1d download
tokenizer.json 30.7 MB a2619fe1 download
model.safetensors.index.json 193 KB 75ce5c50 download
config.json 18.5 KB 6283397e download
chat_template.jinja 16.9 KB 7fa57f97 download
tokenizer_config.json 2.74 KB de3b2c7e download
.gitattributes 1.53 KB 52373fe2 download
README.md 346 B 2f577994 download
generation_config.json 217 B ed42ae71 download

README current version from Hugging Face

library_name: mlx
tags:

  • mlx
  • oq
  • quantized

Gemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic-oQ4-fp16

This model was quantized using oQ (oMLX v0.3.8) mixed-precision quantization.

Quantization details

  • Model type: gemma4
  • Bits: 4
  • Group size: 64
  • Format: MLX safetensors

README history 1 version

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

  1. 2026-05-09Upload Gemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic-oQ4-fp16 via oMLXca04232346 B
    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.

Get the app Browse the catalog
Open in Abliteration