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

BoyoDiffusion/qwen_abliterated_fp8_e4m3fn.safetensors

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/BoyoDiffusion%2Fqwen_abliterated_fp8_e4m3fn.safetensors"
Response includes
  • classification m1
  • files 3
  • author_summary 1 models
  • readme_text full
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)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · 30-day
0
Likes
3
Model age
10mo ago
created 2025-12-02
Downloads over time
Now0→from0↑0%
00110 on Dec 3, 20250 on Oct 11Dec '25FebAprJunAugOct
Dec 3, 2025 → Oct 11 · 84 snapshots · spans 312 days

Metadata

License
mit
Quantizations
FP8
Tags
license:mit region:us
Total size
3.75 GB
Files
3
Quantizations
2
Registered
2026-08-22 13:56
Last updated on HF
2025-12-03 00:16

Files by quantization

FP8 1 file 3.75 GB
qwen_abliterated_fp8_e4m3fn.safetensors 3.75 GB 23ed0cc7 download
Auxiliary files 2 files 2.53 KB
.gitattributes 1.48 KB a6344aac download
README.md 1.05 KB fb6f0372 download

README current version from Hugging Face


license: mit

Qwen3-4B Abliterated Text Encoder

This is an abliterated version of the Qwen3-4B text encoder, converted to ComfyUI-compatible safetensors format for use with models such as Z-Image.

What is this?

This text encoder has had its refusal mechanisms removed, making it suitable for use cases where the original model's safety filtering interferes with legitimate content generation (such as realistic human anatomy in character training).

Usage

Drop this safetensors file into your ComfyUI text encoders folder and use it as a replacement for the standard Qwen3-4B text encoder in your workflows.

Important Notes

  • This model does not refuse requests, but it also does not contain harmful knowledge
  • It maintains the same underlying capabilities as the original Qwen3-4B
  • Intended for technical/artistic use cases where censorship interferes with legitimate content

License

MIT License (inherited from original Qwen3-4B repository)

Credits

Based on the abliterated Qwen3-4B model from huihui-ai, converted to ComfyUI format.

README history 2 versions

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

  1. 2025-12-03Update README.mdfb7f5911 KB
    Loading...
  2. 2025-12-02initial commit204619221 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