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

InfinimindCreations/gemma-4-26B-A4B-it-uncensored

InfinimindCreations Gemma 26B MoE 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/InfinimindCreations%2Fgemma-4-26B-A4B-it-uncensored"
Response includes
  • classification m-uncensored
  • files 14
  • benchmarks 11 entries
  • author_summary 8 models
  • readme_text full
Try next
GET/models?author=InfinimindCreations
All 8 models by InfinimindCreations
 GET/models?family=gemma&method=m-uncensored
Sibling gemma models · same method
 GET/methods/m-uncensored
What "m-uncensored" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M-U
Primary method

Uncensored (method unknown)

No other method signals detected in this model.
Confidence
LOW
Why this label 3 signals
Weak or ambiguous signals. Best guess based on catalog patterns; treat as tentative and check the evidence below.
  • 'uncensored' in name/tags but no 'abliterated' marker
  • method not identifiable from author declaration alone
  • may be DPO fine-tune, prompt engineering, or unknown technique
How we label → About M-U →
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 · 30-day
17
↑ 1,113% in 90 days
Likes
1
Descendants
2
in 2 direct forks
Model age
6mo ago
created 2026-04-07
Downloads over time
Now376→from31↑1,113%
1414627841131 on Apr 15376 on Sep 9AprMayJunJulAugSep
Apr 15 → Sep 9 · 33 snapshots · spans 147 days

Benchmarks

Portrait before abliteration
Benchmarks of the base model as it stood before the refusal-removal operation. Compare with the numbers above to see what the operation cost.
Benchmark Score Source
Entertainment 2.2 UGI
Hazardous 2.9 UGI
Natural Intelligence 34.44 UGI
Political lean -18.2% UGI
Sensitive-Info 22.41 UGI
SocPol 1.8 UGI
UGI 20.77 UGI
Willingness (10) 1.8 UGI
W10-Adherence 1.5 UGI
W10-Direct 2 UGI
Writing 41.62 UGI

Genealogy 2 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-4-26B-A4B-it this lineage
InfinimindCreations/gemma-4-26B-A4B-it-uncensored↓ 17 2 forks
mradermacher/gemma-4-26B-A4B-it-uncensored-i1-GGUF↓ 7,115
mradermacher/gemma-4-26B-A4B-it-uncensored-GGUF↓ 993

Metadata

License
gemma
Tags
transformers safetensors gemma4 image-text-to-text abliteration uncensored biprojection moe conversational arxiv:2406.11717 base_model:google/gemma-4-26B-A4B-it base_model:finetune:google/gemma-4-26B-A4B-it

Related

Other by InfinimindCreations
Spark-X2.5-4B-uncensored-GGUF 6K K2-Horizon-MoVA-36B-A4B-uncensored-GGUF 3K gemma-4-E4B-it-uncensored-vision 2K gemma-4-E4B-it-uncensored 954 gemma-4-31B-it-uncensored 638
Same base model
HauhauCS/Gemma4-26B-A4B-QAT-Uncensored-HauhauCS-Balanced-MTP 411K HauhauCS/Gemma4-26B-A4B-Uncensored-HauhauCS-Balanced 72K Jiunsong/supergemma4-26b-uncensored-gguf-v2 48K Jiunsong/supergemma4-26b-uncensored-mlx-4bit-v2 9K OS-Software/gemma-4-26B-A4B-it-qat-q4_0-uncensored-heretic-v2-GGUF 3K
Total size
48.1 GB
Files
14
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-04-07 18:59
Source huggingface.co/InfinimindCreations/gemma-4-26B-A4B-it-uncensored ↗ Base model google/gemma-4-26B-A4B-it ↗

Files by quantization

Auxiliary files 14 files 48.1 GB
model-00003-of-00006.safetensors 9.13 GB 96eb745c download
model-00004-of-00006.safetensors 9.13 GB c275c91b download
model-00005-of-00006.safetensors 9.13 GB b8330811 download
model-00002-of-00006.safetensors 9.13 GB d5ff6dad download
model-00001-of-00006.safetensors 8.96 GB 66a9ca01 download
model-00006-of-00006.safetensors 2.60 GB 250af069 download
tokenizer.json 30.7 MB be504a0a download
model.safetensors.index.json 101 KB 6f9e80d8 download
chat_template.jinja 11.8 KB 33c51c2d download
config.json 3.72 KB 5b0b791b download
tokenizer_config.json 2.62 KB f07b8ede download
.gitattributes 1.53 KB 52373fe2 download
README.md 1.35 KB 264c329a download
generation_config.json 203 B edda3c19 download

README current version from Hugging Face

license: gemma
base_model: google/gemma-4-26B-A4B-it
tags:

  • abliteration
  • uncensored
  • gemma4
  • biprojection
  • moe
    library_name: transformers

Gemma 4 26B-A4B-it Uncensored (Biprojection + EGA)

Uncensored version of google/gemma-4-26B-A4B-it via biprojection abliteration with
Expert-Granular Abliteration (EGA) for MoE expert weights.

Method

  • Norm-preserving biprojection (double Gram-Schmidt + double projection pass)
  • Winsorization 0.995, 400+400 prompts (mlabonne/harmful_behaviors + tatsu-lab/alpaca)
  • Dense pathway: o_proj + mlp.down_proj (60 weights)
  • MoE experts: all 128 experts per layer (3840 weights)
  • Total: 3900 weights modified across 30 layers
  • Precision: bf16, no quantization during abliteration

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("InfinimindCreations/gemma-4-26B-A4B-it-uncensored", torch_dtype="bfloat16", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("InfinimindCreations/gemma-4-26B-A4B-it-uncensored")

Credits

README history 1 version

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

  1. 2026-04-07Biprojection+EGA abliteration (bf16, 3900 weights, 30 layers)4b7789b1.3 KB
    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