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

smashingtags/gemma-4-12b-it-abliterated

smashingtags Gemma 12B multimodal
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/smashingtags%2Fgemma-4-12b-it-abliterated"
Response includes
  • classification m1
  • files 9
  • hub_downloads_all_time 35
  • author_summary 3 models
  • readme_text full
Try next
GET/models?author=smashingtags
All 3 models by smashingtags
 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
35
18 last 30d - active
Likes
0
Model age
3mo ago
created 2026-06-14
Downloads over time
Now45→from5↑800%
31834495 on Jul 145 on Oct 1145 on Oct 9JulAugSepOct
Jul 1 → Oct 11 · 54 snapshots · spans 102 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.

google/gemma-4-12B-it-qat-q4_0-unquantized this lineage
smashingtags/gemma-4-12b-it-abliterated↓ 18

Variants by this author 2 formats · 72 downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

License
gemma
Tags
safetensors gemma4_unified abliterated uncensored gemma-4 multimodal base_model:google/gemma-4-12B-it-qat-q4_0-unquantized base_model:finetune:google/gemma-4-12B-it-qat-q4_0-unquantized license:gemma region:us

Related

Other by smashingtags
gemma-4-12b-it-abliterated-nf4 102 gemma-4-12b-it-abliterated-GGUF 54
Same base model
huihui-ai/Huihui-gemma-4-12B-it-qat-q4_0-unquantized-abliterated-GGUF 32K huihui-ai/Huihui-gemma-4-12B-it-qat-q4_0-unquantized-abliterated 1K llmfan46/gemma-4-12B-it-qat-q4_0-unquantized-uncensored-heretic 296 tinybiggames/Huihui-gemma-4-12B-it-qat-q4_0-unquantized-abliterated-GGUF 175 BigBen86/Huihui-gemma-4-12B-it-qat-q4_0-unquantized-abliterated-GGUF 115
Total size
22.3 GB
Files
9
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-06-14 17:32
Source huggingface.co/smashingtags/gemma-4-12b-it-abliterated ↗ Base model google/gemma-4-12B-it-qat-q4_0-unquantized ↗

Files by quantization

Auxiliary files 9 files 22.3 GB
model.safetensors 22.3 GB 35391191 download
tokenizer.json 30.7 MB cc8d3a0c download
chat_template.jinja 17.1 KB e61bbfe9 download
config.json 4.20 KB fdecc804 download
tokenizer_config.json 2.69 KB 970d3843 download
.gitattributes 1.53 KB 52373fe2 download
processor_config.json 1.35 KB b889adcd download
README.md 519 B 04354a0f download
generation_config.json 255 B 65fd0cf2 download

README current version from Hugging Face

base_model: google/gemma-4-12B-it-qat-q4_0-unquantized
license: gemma
tags: [abliterated, uncensored, gemma-4, multimodal]

gemma-4-12b-it-abliterated

Abliterated (refusal-direction-removed) build of Google's official QAT
gemma-4-12B-it-qat-q4_0-unquantized. Text + image + audio retained — only the
text decoder's o_proj/mlp.down_proj (top 70% of layers) were edited; vision
and audio encoders and the tied embeddings are untouched. Refusal smoke test:
3/3 (base) -> 0/3 (this model). Personal use.

README history 1 version

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

  1. 2026-06-14EOS-225: model cardb9eda38519 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