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

kanarena/supergemma4-e4b-abliterated-multimodal-gguf-4bit

kanarena Gemma GGUF multimodal second-order 131K ctx
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/kanarena%2Fsupergemma4-e4b-abliterated-multimodal-gguf-4bit"
Response includes
  • classification m8
  • files 4
  • hub_downloads_all_time 3,381
  • author_summary 1 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

Applied on top of direct removal inherited from the base model.
Confidence
MEDIUM
Inherited from base model
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=1
  • assume M1 (base ablation) + M8 (GGUF quant) - default when producer unknown
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · lifetime
3K
287 last 30d - cooling
Likes
2
Model age
5mo ago
created 2026-04-19
Downloads over time
Now3.5K→from779↑345%
6451.7K2.7K3.7K779 on Apr 223.5K on Oct 11AprMayJunJulAugSepOct
Apr 22 → Oct 11 · 64 snapshots · spans 172 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
gemma
Languages
en ko
Quantizations
Q4_K
Tags
gguf gemma4 llama.cpp multimodal image-text-to-text abliterated uncensored quantized 4-bit en ko base_model:Jiunsong/supergemma4-e4b-abliterated

Related

Total size
4.97 GB
Files
4
Quantizations
3
Registered
2026-08-22 13:56
Last updated on HF
2026-04-19 04:33

Files by quantization

Q4_K 1 file 4.97 GB
Supergemma4-E4B-Abliterated-7.5B-Q4_K_M.gguf 4.97 GB dada5fe9 download
BF16 1 file 946 MB
mmproj-gemma-4-E4B-it-BF16.gguf 946 MB 68486d68 download
Auxiliary files 2 files 5.43 KB
README.md 3.80 KB c052f8b0 download
.gitattributes 1.63 KB 9b221050 download

README current version from Hugging Face


license: gemma
base_model:

  • Jiunsong/supergemma4-e4b-abliterated
    tags:
    • gemma4
    • gguf
    • llama.cpp
    • multimodal
    • image-text-to-text
    • abliterated
    • uncensored
    • quantized
    • 4-bit
      language:
    • en
    • ko
      pipeline_tag: image-text-to-text

SuperGemma4-E4B-Abliterated-Multimodal GGUF 4bit

This is the llama.cpp-ready GGUF 4bit distribution of Jiunsong/supergemma4-e4b-abliterated.

Contents include:

How to Install

It is a installation guide for using the model in LM Studio on Windows.
Before proceeding with these steps, you must first install and launch LM Studio."

Instructions:

  1. Open Windows Terminal (PowerShell).

  2. Change the directory to LM Studio's "Models Directory." (If you specified a different path for your Models Directory, change to that custom path instead.)

cd $env:USERPROFILE\.lmstudio\models
  1. Clone the following repository into a folder named kanarena.
git clone https://huggingface.co/kanarena/supergemma4-e4b-abliterated-multimodal-gguf-4bit .\kanarena\supergemma4-e4b-abliterated-multimodal-gguf-4bit\
  1. Move the configuration files (virtual model) to the main models directory.
Move-Item -Path ".\kanarena\supergemma4-e4b-abliterated-multimodal-gguf-4bit\jiunsong" -Destination ".\"
  1. You can now use this model in LM Studio. If the model does not appear, please Refresh model list.

SuperGemma4-E4B-Abliterated-Multimodal GGUF 4bit

이 배포판은 Jiunsong/supergemma4-e4b-abliterated 모델을 llama.cpp 또는 LM 스튜디오 에서 사용할 수 있는 GGUF 파일입니다.

  • 변환 및 양자화 모델: Jiunsong/supergemma4-e4b-abliterated에서 변환되고 양자화된 Jiunsong/supergemma4-e4b-abliterated 모델 파일.
  • 이미지 기능용 모델 (mmproj): 비전(Vision) 기능을 활성화하기 위해 lmstudio-community/gemma-4-E4B-it-GGUF에서 가져온 이미지 관련 mmproj 파일.
  • 추론 설정 파일 (가상 모델): LM Studio 내에서 추론(Reasoning, Think로 알려져있음) 기능 사용을 위한 가상 모델 설정 파일.

설치법

이 설치법은 윈도우용 LM 스튜디오에서 모델 사용하기 위한 것입니다.
아래 설치법을 따라하기 전에 먼저 LM 스튜디오를 설치하고 최초 실행을 하여야 합니다.

설치법:

  1. 윈도우 터미널(Powershell)을 엽니다.

  2. LM Studio의 "Models Directory" 위치로 디렉토리를 변경합니다. (만약 Models Directory를 다른 경로로 지정했다면 해당 경로로 디렉토리를 변경합니다.)

cd $env:USERPROFILE\.lmstudio\models
  1. 이 레포지토리를 kanarena 이름의 폴더 안에 다운로드 합니다.
git clone https://huggingface.co/kanarena/supergemma4-e4b-abliterated-multimodal-gguf-4bit .\kanarena\supergemma4-e4b-abliterated-multimodal-gguf-4bit\
  1. 설정 파일(가상 모델)을 model(상위 디렉토리)로 이동합니다.
Move-Item -Path ".\kanarena\supergemma4-e4b-abliterated-multimodal-gguf-4bit\jiunsong" -Destination ".\"
  1. 이제 LM Studio 에서 이 모델을 사용할 수 있습니다. 만약 보이지 않는 경우 모델 리스트 새로고침을 해주세요.

README history 3 versions

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

  1. 2026-04-19Update README.md60ee6693.8 KB
    Loading...
  2. 2026-04-19Update README.md6c74efe3.5 KB
    Loading...
  3. 2026-04-19Initial commit29de967263 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.

Open in Abliteration