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

EZCon/Huihui-Qwen3-VL-4B-Instruct-abliterated-4bit-g32-mxfp4-mixed_4_8-mlx

EZCon Qwen 4.0B 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/EZCon%2FHuihui-Qwen3-VL-4B-Instruct-abliterated-4bit-g32-mxfp4-mixed_4_8-mlx"
Response includes
  • classification m1
  • files 16
  • hub_downloads_all_time 3,692
  • author_summary 31 models
  • readme_text full
Try next
GET/models?author=EZCon
All 31 models by EZCon
 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
4K
182 last 30d - cooling
Likes
1
Model age
8mo ago
created 2026-01-29
Downloads over time
Now3.8K→from13↑28,885%
01.4K2.8K4.1K13 on Jan 283.8K on Oct 11JanMarMayJulSep
Jan 28 → Oct 11 · 76 snapshots · spans 256 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.

huihui-ai/Huihui-Qwen3-VL-4B-Instruct-abliterated↓ 15,140 this lineage
EZCon/Huihui-Qwen3-VL-4B-Instruct-abliterated-4bit-g32-mxfp4-mixed_4_8-mlx↓ 182

Metadata

License
apache-2.0
Tags
mlx safetensors qwen3_vl abliterated uncensored image-text-to-text conversational base_model:huihui-ai/Huihui-Qwen3-VL-4B-Instruct-abliterated base_model:quantized:huihui-ai/Huihui-Qwen3-VL-4B-Instruct-abliterated license:apache-2.0 4-bit region:us

Related

Other by EZCon
Huihui-Qwen3-VL-4B-Instruct-abliterated-mlx 500 Huihui-gemma-4-E4B-it-abliterated-4bit-mlx 422 Qwen2-VL-2B-Instruct-abliterated-mlx 389 Huihui-Qwen3-VL-4B-Instruct-abliterated-8bit-mlx 379 Huihui-Qwen3-VL-2B-Instruct-abliterated-mlx 336
Same base model
noctrex/Huihui-Qwen3-VL-4B-Instruct-abliterated-GGUF 28K mradermacher/Huihui-Qwen3-VL-4B-Instruct-abliterated-GGUF 3K noctrex/Huihui-Qwen3-VL-4B-Instruct-abliterated-i1-GGUF 1K mradermacher/Huihui-Qwen3-VL-4B-Instruct-abliterated-i1-GGUF 1K EZCon/Huihui-Qwen3-VL-4B-Instruct-abliterated-mlx 500
Total size
3.29 GB
Files
16
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-08-12 11:52
Source huggingface.co/EZCon/Huihui-Qwen3-VL-4B-Instruct-abliterated-4bit-g32-mxfp4-mixed_4_8-mlx ↗ Base model huihui-ai/Huihui-Qwen3-VL-4B-Instruct-abliterated →

Files by quantization

Auxiliary files 16 files 3.31 GB
model.safetensors 3.29 GB cf90d734 download
tokenizer.json 10.9 MB be756060 download
vocab.json 2.65 MB 4783fe10 download
merges.txt 1.59 MB 31349551 download
model.safetensors.index.json 97.6 KB da24e2f1 download
config.json 62.7 KB 1cb67f62 download
chat_template.jinja 5.17 KB 12438680 download
.gitattributes 1.53 KB 52373fe2 download
processor_config.json 989 B d0bf62ac download
README.md 872 B d628130f download
video_preprocessor_config.json 817 B e32b1d90 download
preprocessor_config.json 782 B 2fa65535 download
added_tokens.json 707 B b54f9135 download
special_tokens_map.json 613 B ac23c0aa download
tokenizer_config.json 418 B e103cbdf download
generation_config.json 213 B bdb4e037 download

README current version from Hugging Face

license: apache-2.0
base_model: huihui-ai/Huihui-Qwen3-VL-4B-Instruct-abliterated
pipeline_tag: image-text-to-text
tags:

  • abliterated
  • uncensored
  • mlx
    library_name: mlx

EZCon/Huihui-Qwen3-VL-4B-Instruct-abliterated-4bit-g32-mxfp4-mixed_4_8-mlx

This model was converted to MLX format from huihui-ai/Huihui-Qwen3-VL-4B-Instruct-abliterated
using mlx-vlm version 0.6.12.
Refer to the original model card for more details on the model.

Use with mlx

pip install -U mlx-vlm

python -m mlx_vlm.generate --model EZCon/Huihui-Qwen3-VL-4B-Instruct-abliterated-4bit-g32-mxfp4-mixed_4_8-mlx --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>

README history 14 versions

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

  1. 2026-08-12Upload folder using huggingface_hub9a6c06c872 B
    Loading...
  2. 2026-07-15Upload folder using huggingface_hub774a11c871 B
    Loading...
  3. 2026-07-08Upload folder using huggingface_hub79afba4871 B
    Loading...
  4. 2026-06-12Upload folder using huggingface_hub22b731a871 B
    Loading...
  5. 2026-06-03Upload folder using huggingface_hub28c5ef9871 B
    Loading...
  6. 2026-05-19Upload folder using huggingface_hubd11859d871 B
    Loading...
  7. 2026-04-04Upload folder using huggingface_hub05b2569871 B
    Loading...
  8. 2026-04-03Upload folder using huggingface_hub68a219e871 B
    Loading...
  9. 2026-03-29Upload folder using huggingface_hubd4425ce871 B
    Loading...
  10. 2026-03-24Upload folder using huggingface_hub2758804784 B
    Loading...
  11. 2026-03-12Upload folder using huggingface_hub7eb25a2784 B
    Loading...
  12. 2026-02-23Upload folder using huggingface_hub29050bf785 B
    Loading...
  13. 2026-02-05Upload folder using huggingface_hub49c9a91785 B
    Loading...
  14. 2026-01-29Upload folder using huggingface_hubf50ae8b785 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