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

ArtusDev/grimjim_gemma-3-12b-it-projection-abliterated-EXL3

ArtusDev 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/ArtusDev%2Fgrimjim_gemma-3-12b-it-projection-abliterated-EXL3"
Response includes
  • classification m1
  • files 2
  • benchmarks 11 entries
  • hub_downloads_all_time 83
  • author_summary 8 models
  • readme_text full
Try next
GET/models?author=ArtusDev
All 8 models by ArtusDev
 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
83
21 last 30d - stable
Likes
1
Model age
11mo ago
created 2025-10-30
Downloads over time
Now88→from9↑878%
53566969 on Oct 29, 202588 on Oct 1288 on Oct 10Oct '25Dec '25FebAprJunAugOct
Oct 29, 2025 → Oct 12 · 90 snapshots · spans 348 days

Benchmarks

Benchmark Score Source
Entertainment 0.9 UGI
Hazardous 1.2 UGI
Natural Intelligence 19.21 UGI
Political lean -6.9% UGI
Sensitive-Info 9.92 UGI
SocPol 1 UGI
UGI 39.11 UGI
Willingness (10) 9.8 UGI
W10-Adherence 9.5 UGI
W10-Direct 10 UGI
Writing 29.46 UGI

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.

grimjim/gemma-3-12b-it-projection-abliterated↓ 63 this lineage
ArtusDev/grimjim_gemma-3-12b-it-projection-abliterated-EXL3↓ 21

Metadata

License
gemma
Tags
transformers exl3 image-text-to-text base_model:grimjim/gemma-3-12b-it-projection-abliterated base_model:quantized:grimjim/gemma-3-12b-it-projection-abliterated license:gemma endpoints_compatible region:us

Related

Other by ArtusDev
huihui-ai_Qwen2.5-Coder-32B-Instruct-abliterated-EXL3 39 huihui-ai_Huihui-Qwen3-30B-A3B-Thinking-2507-abliterated-EXL3 35 huihui-ai_Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-EXL3 34 huihui-ai_Qwen3-30B-A3B-abliterated_EXL3_4.0bpw_H6 27 huihui-ai_Qwen2.5-Coder-32B-Instruct-abliterated-EXL2 23
Same base model
mradermacher/gemma-3-12b-it-projection-abliterated-i1-GGUF 1K mradermacher/gemma-3-12b-it-projection-abliterated-GGUF 508 CallMcMargin/gemma-3-12b-it-projection-abliterated-mlx-bf16-mxfp4-mixed-4-6 44
Total size
0 B
Files
2
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-10-30 08:17
Source huggingface.co/ArtusDev/grimjim_gemma-3-12b-it-projection-abliterated-EXL3 ↗ Base model grimjim/gemma-3-12b-it-projection-abliterated →

Files by quantization

Auxiliary files 2 files 12.4 KB
README.md 10.9 KB b04f21f3 download
.gitattributes 1.48 KB a6344aac download

README current version from Hugging Face

base_model: grimjim/gemma-3-12b-it-projection-abliterated
base_model_relation: quantized
quantized_by: ArtusDev
license: gemma
library_name: transformers
pipeline_tag: image-text-to-text
tags:

  • exl3

ArtusDev/grimjim_gemma-3-12b-it-projection-abliterated-EXL3

EXL3 quants of grimjim/gemma-3-12b-it-projection-abliterated using exllamav3 for quantization.

Quants

Quant BPW Head Bits Size (GB)
2.5_H6 2.5 6 7.02
3.0_H6 3.0 6 7.70
3.5_H6 3.5 6 8.37
4.0_H6 4.0 6 9.04
4.5_H6 4.5 6 9.71
5.0_H6 5.0 6 10.39
6.0_H6 6.0 6 11.73
8.0_H8 8.0 8 14.67

How to Download and Use Quants

You can download quants by targeting specific size using the Hugging Face CLI.

Click for download commands
1. Install huggingface-cli: 
pip install -U "huggingface_hub[cli]"
2. Download a specific quant: 
huggingface-cli download ArtusDev/grimjim_gemma-3-12b-it-projection-abliterated-EXL3 --revision "5.0bpw_H6" --local-dir ./

EXL3 quants can be run with any inference client that supports EXL3, such as TabbyAPI. Refer to documentation for set up instructions.

Quant Requests

See EXL community hub for request guidelines.

Acknowledgements

Made possible with cloud compute from lium.io

README history 10 versions

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

  1. 2025-10-30Update README.md661370c10.9 KB
    Loading...
  2. 2025-10-30Upload README.md with huggingface_hubb9e6ed710.6 KB
    Loading...
  3. 2025-10-30Upload folder using huggingface_hubd6320f5727 B
    Loading...
  4. 2025-10-30Upload folder using huggingface_hub0cec055727 B
    Loading...
  5. 2025-10-30Upload folder using huggingface_hub16eebbb727 B
    Loading...
  6. 2025-10-30Upload folder using huggingface_hub4ee6ba0727 B
    Loading...
  7. 2025-10-30Upload folder using huggingface_hub7b9987d727 B
    Loading...
  8. 2025-10-30Upload folder using huggingface_hubeb50b2c727 B
    Loading...
  9. 2025-10-30Upload folder using huggingface_hubfb6f3ac727 B
    Loading...
  10. 2025-10-30Upload folder using huggingface_hub56ae1ed727 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