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

fsxedx/Qwen3-VL-4B-Instruct-Uncensored-abliterated

fsxedx Qwen 4.4B multimodal second-order
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/fsxedx%2FQwen3-VL-4B-Instruct-Uncensored-abliterated"
Response includes
  • classification m1
  • files 11
  • hub_downloads_all_time 126
  • author_summary 6 models
  • readme_text full
Try next
GET/models?author=fsxedx
All 6 models by fsxedx
 GET/models?family=qwen&method=m1
Sibling qwen 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 · lifetime
126
40 last 30d - stable
Likes
0
Model age
2mo ago
created 2026-08-05
Downloads over time
Now135→from10↑1,250%
45210014810 on Aug 5135 on Oct 11135 on Oct 10AugSepOct
Aug 5 → Oct 11 · 50 snapshots · spans 67 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.

Felldude/Qwen3-VL-4B-Instruct-Uncensored↓ 151 this lineage
fsxedx/Qwen3-VL-4B-Instruct-Uncensored-abliterated↓ 40

Metadata

License
apache-2.0
Tags
safetensors qwen3_vl heretic abliterated caption image-text-to-text conversational base_model:Felldude/Qwen3-VL-4B-Instruct-Uncensored base_model:merge:Felldude/Qwen3-VL-4B-Instruct-Uncensored base_model:Felldude/Qwen3-VL-4B-Instruct-Uncensored-FP8 base_model:merge:Felldude/Qwen3-VL-4B-Instruct-Uncensored-FP8 license:apache-2.0

Related

Other by fsxedx
Qwen3-4b-Z-Image-Turbo-AbliteratedV1 489 Llama-3.2-4X3B-MOE-Hell-California-Uncensored-10B-GGUF 285 Qwen3-4b-Z-Image-Turbo-AbliteratedV1-GGUF 233 Llama-3.2-8X3B-MOE-Dark-Champion-Instruct-uncensored-abliterated-18.4B-GGUF 202 Qwen3-VL-4B-Instruct-Uncensored-FP8 16
Same base model
mradermacher/Qwen3-VL-4B-Instruct-Uncensored-GGUF 2K mradermacher/Qwen3-VL-4B-Instruct-Uncensored-i1-GGUF 1K dummy9996/Qwen3-VL-4B-Instruct-Uncensored-abliterated 168
Total size
8.27 GB
Files
11
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-08-05 18:01
Source huggingface.co/fsxedx/Qwen3-VL-4B-Instruct-Uncensored-abliterated ↗ Base model Felldude/Qwen3-VL-4B-Instruct-Uncensored →

Files by quantization

Auxiliary files 11 files 8.28 GB
model.safetensors 8.27 GB 395a6853 download
tokenizer.json 6.71 MB c6cc1014 download
vocab.json 2.65 MB 4783fe10 download
tokenizer_config.json 10.6 KB d3d37632 download
chat_template.json 5.37 KB e49e75bd download
config.json 1.52 KB 36c5f35d download
.gitattributes 1.48 KB a6344aac download
preprocessor_config.json 390 B 2ea84a43 download
video_preprocessor_config.json 385 B 3ba673a5 download
README.md 355 B 25140882 download
generation_config.json 269 B e116347d download

README current version from Hugging Face

base_model:

  • Felldude/Qwen3-VL-4B-Instruct-Uncensored
  • Felldude/Qwen3-VL-4B-Instruct-Uncensored-FP8
    base_model_relation: merge
    pipeline_tag: image-text-to-text
    tags:
  • heretic
  • abliterated
  • caption
    license: apache-2.0

have abliterated the decensored captioning finetune, won't refuse anymore


@Felldude

README history 1 version

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

  1. 2026-08-05Duplicate from dummy9996/Qwen3-VL-4B-Instruct-Uncensored-abliterated09c3263355 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.

Get the app Browse the catalog
Open in Abliteration