Get the app
← back to catalog · registered 2026-09-25 04:57

pekkAi/Gemma-4-E2B-E4B-Abliterated-Text-Encoders-ComfyUI

pekkAi Gemma second-order 
curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/pekkAi%2FGemma-4-E2B-E4B-Abliterated-Text-Encoders-ComfyUI"
Response includes
  • classification m1
  • files 6
  • author_summary 10 models
  • readme_text full
Try next
GET/models?author=pekkAi
All 10 models by pekkAi
 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 · 30-day
0
Likes
0
Model age
today
created 2026-09-25

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.

huihui-ai/Huihui-gemma-4-E2B-it-abliterated↓ 2,058 this lineage
pekkAi/Gemma-4-E2B-E4B-Abliterated-Text-Encoders-ComfyUI

Metadata

License
apache-2.0
Quantizations
BF16 FP8
Tags
comfyui text-encoder fp8 base_model:huihui-ai/Huihui-gemma-4-E2B-it-abliterated base_model:finetune:huihui-ai/Huihui-gemma-4-E2B-it-abliterated license:apache-2.0 region:us

Related

Other by pekkAi
Gemma-4-Queen-31B-it-uncensored-heretic-NVFP4 919 G4-MeroMero-26B-A4B-it-uncensored-heretic-NVFP4 297 gemma-4-E4B-it-ultra-uncensored-heretic-nvfp4 290 Qwen3.5-9B-Uncensored-HauhauCS-Aggressive-NVFP4 188 Gemma-4-Gembrain-31B-it-uncensored-heretic-NVFP4 136
Same base model
mradermacher/Huihui-gemma-4-E2B-it-abliterated-GGUF 3K mradermacher/Huihui-gemma-4-E2B-it-abliterated-i1-GGUF 1K edp1096/Huihui-gemma-4-E2B-it-abliterated-litert-lm 325 EZCon/Huihui-gemma-4-E2B-it-abliterated-4bit-g32-mxfp4-mixed_4_8-mlx 138 EZCon/Huihui-gemma-4-E2B-it-abliterated-8bit-mlx 130
Total size
38.4 GB
Files
6
Quantizations
3
Registered
2026-09-25 04:57
Last updated on HF
2026-09-25 04:02
Source huggingface.co/pekkAi/Gemma-4-E2B-E4B-Abliterated-Text-Encoders-ComfyUI ↗ Base model huihui-ai/Huihui-gemma-4-E2B-it-abliterated →

Files by quantization

BF16 2 files 24.5 GB
gemma4_e4b_abliterated_bf16.safetensors 14.9 GB 021ce712 download
gemma4_e2b_abliterated_bf16.safetensors 9.57 GB 805cbcb3 download
FP8 2 files 13.9 GB
gemma4_e4b_abliterated_fp8_scaled.safetensors 8.44 GB 23109821 download
gemma4_e2b_abliterated_fp8_scaled.safetensors 5.47 GB 08a9d2ac download
Auxiliary files 2 files 2.41 KB
.gitattributes 1.54 KB 89f551a4 download
README.md 895 B 056f0f3d download

README current version from Hugging Face

license: apache-2.0
tags:

  • comfyui
  • text-encoder
  • fp8
    base_model:
  • huihui-ai/Huihui-gemma-4-E2B-it-abliterated
  • huihui-ai/Huihui-gemma-4-E4B-it-abliterated

Gemma 4 - E2B/E4B - abliterated - text encoders

Repackaged models for usage as text encoders in ComfyUI

Original model repository:

Place the encoder files in the following folder:

📂 ComfyUI/
└── 📂 models/
    └── 📂 text_encoders/
        ├── gemma4_e2b_heretic_bf16.safetensors
        ├── gemma4_e2b_heretic_fp8_scaled.safetensors
        ├── gemma4_e4b_heretic_bf16.safetensors
        └── gemma4_e4b_heretic_fp8_scaled.safetensors

README history 1 version

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

  1. 2026-09-25Upload folder using huggingface_hub99489bd895 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 app" button that hands off directly to a local runtime of your choice - Abliteration, LM Studio, or Ollama. No API keys, no subscription, no prompt leakage.

See supported apps Browse the catalog