← back to catalog · registered 2026-09-15 19:56

lucifer053/gemma-4-26b-a4b-it-uncensored-th-gguf

lucifer053 Gemma 26B GGUF MoE
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 · 30-day
0
Likes
0
Model age
today
created 2026-09-15

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
th en
Quantizations
Q3_K Q4_K
Tags
gguf abliterated uncensored thai heretic moe text-generation th en license:gemma endpoints_compatible region:us

Related

Total size
28.0 GB
Files
4
Quantizations
3
Registered
2026-09-15 19:56
Last updated on HF
2026-09-15 20:08

Files by quantization

Q4_K 1 file 15.6 GB
gemma-4-26B-A4B-it-heretic-th-Q4_K_M.gguf 15.6 GB 1b2597f5 download
Q3_K 1 file 12.4 GB
gemma-4-26B-A4B-it-heretic-th-Q3_K_M.gguf 12.4 GB dcb597da download
Auxiliary files 2 files 3.61 KB
README.md 1.97 KB d9afd77b download
.gitattributes 1.64 KB d00910ef download

README current version from Hugging Face


license: gemma
base_model: google/gemma-4-26b-a4b-it
tags:

  • abliterated
  • uncensored
  • gguf
  • thai
  • heretic
  • moe
    language:
  • th
  • en
    pipeline_tag: text-generation

Gemma 4 26B A4B Uncensored TH (GGUF)

An abliterated (refusal-direction-removed) derivative of
google/gemma-4-26b-a4b-it.

This is a modified model derivative created under the Gemma Terms of Use.

What makes this different

Most abliterated models are measured with English-only refusal datasets,
which often degrades output quality in other languages. This model was
abliterated using a Thai prompt dataset (with English refusal markers kept
as a fallback), so it stays fluent in Thai while reducing over-refusals.

This upload is text GGUF only. Vision/mmproj is not included here.

Files

File Quant Size (approx.) Use case
gemma-4-26B-A4B-it-heretic-th-Q5_K_M.gguf Q5_K_M 16–18 GB recommended
gemma-4-26B-A4B-it-heretic-th-Q4_K_M.gguf Q4_K_M 13–15 GB 32 GB VRAM
gemma-4-26B-A4B-it-heretic-th-Q3_K_M.gguf Q3_K_M 11–13 GB smaller footprint

Usage

llama.cpp

./llama-cli \
  -m gemma-4-26B-A4B-it-heretic-th-Q5_K_M.gguf \
  -ngl 99 \
  --reasoning off --reasoning-budget 0 \
  -c 8192 \
  --temp 0.7 \
  --min-p 0.05 \
  --top-p 0.9 \
  --repeat-penalty 1.12 \
  --repeat-last-n 256

Do not drop the temperature below ~0.6. Abliterated models tend to loop at
very low temperature. For longer writing, --temp 0.78 and -c 16384 work
better than raising --repeat-penalty.

LM Studio

Load the GGUF directly. Turn thinking/reasoning off if the UI exposes that
toggle.

How it was made

Base model: google/gemma-4-26b-a4b-it (MoE, 26B total / 4B active)
Method: abliteration via Heretic
Prompt dataset: Thai harmful / harmless pairs used for measurement and
evaluation, so Thai output stays natural after the edit
Selected trial: 199 (0/16 refusals, KL 0.0074)

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 app" button that hands off directly to a local runtime of your choice - Infrahuman, LM Studio, or Ollama. No API keys, no subscription, no prompt leakage.