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

stronman/gemma-4-31b-abliterated-mlx-4bit

stronman Gemma 3.8B
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/stronman%2Fgemma-4-31b-abliterated-mlx-4bit"
Response includes
  • classification m1
  • files 13
  • hub_downloads_all_time 48
  • author_summary 6 models
  • readme_text full
Try next
GET/models?author=stronman
All 6 models by stronman
 GET/models?family=gemma&method=m1
Sibling gemma models · same method
 GET/methods/m1
What "m1" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
How we label → About M1 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
48
9 last 30d - stable
Likes
0
Model age
6mo ago
created 2026-04-11
Downloads over time
Now49→from18↑172%
1628405218 on Apr 1549 on Oct 1149 on Oct 1AprMayJunJulAugSepOct
Apr 15 → Oct 11 · 65 snapshots · spans 179 days

Metadata

License
apache-2.0
Tags
safetensors gemma4 broken wip license:apache-2.0 4-bit region:us

Related

Other by stronman
Gemma-4-E4B-Uncensored-HauhauCS-Aggressive 554 Gemma-4-31B-it-abliterated-GGUF 432 Qwen3.5-35B-A3B-Uncensored-HauhauCS-Aggressive 173 Gemma-4-31B-it-abliterated 60 gemma-4-31B-Claude-Opus-Distill-abliterated-mlx-mxfp8 51
Total size
17.1 GB
Files
13
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-04-11 04:04
Source huggingface.co/stronman/gemma-4-31b-abliterated-mlx-4bit ↗

Files by quantization

Auxiliary files 13 files 17.2 GB
model-00003-of-00004.safetensors 5.00 GB 7bd80c3e download
model-00001-of-00004.safetensors 5.00 GB fc6407fb download
model-00002-of-00004.safetensors 4.99 GB 6fcb9459 download
model-00004-of-00004.safetensors 2.16 GB 4083f551 download
tokenizer.json 30.7 MB cc8d3a0c download
model.safetensors.index.json 201 KB 035c187f download
chat_template.jinja 11.8 KB 33c51c2d download
config.json 5.51 KB d3e15453 download
tokenizer_config.json 2.62 KB 59dd4b62 download
.gitattributes 1.53 KB 52373fe2 download
processor_config.json 627 B 09dfe230 download
README.md 326 B 3684e5a3 download
generation_config.json 208 B e605bb45 download

README current version from Hugging Face

license: apache-2.0
tags:

  • broken
  • wip

WARNING: This model has degeneration issues

This model is under active development. The current weights cause text degeneration (repetition loops) after ~200-300 tokens. Do not use in production.

A fixed version with gentler abliteration parameters is being developed.

README history 1 version

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

  1. 2026-04-11Duplicate from dboris/gemma-4-31b-abliterated-mlx-4bit309cd03326 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