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

AITRADER/Qwen2.5-VL-32B-Instruct-abliterated-mlx-8bit

AITRADER Qwen 33B multimodal
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/AITRADER%2FQwen2.5-VL-32B-Instruct-abliterated-mlx-8bit"
Response includes
  • classification m1
  • files 21
  • benchmarks 11 entries
  • hub_downloads_all_time 1,304
  • author_summary 30 models
  • readme_text full
Try next
GET/models?author=AITRADER
All 30 models by AITRADER
 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
1K
92 last 30d - cooling
Likes
0
Model age
7mo ago
created 2026-02-15
Downloads over time
Now1.3K→from162↑725%
1035541K1.5K162 on Feb 181.3K on Oct 11FebAprJunAugOct
Feb 18 → Oct 11 · 73 snapshots · spans 235 days

Benchmarks

Portrait before abliteration
Benchmarks of the base model as it stood before the refusal-removal operation. Compare with the numbers above to see what the operation cost.
Benchmark Score Source
Entertainment 0.8 UGI
Hazardous 2.4 UGI
Natural Intelligence 16.02 UGI
Political lean -17.4% UGI
Sensitive-Info 17.99 UGI
SocPol 2.6 UGI
UGI 16.99 UGI
Willingness (10) 1.5 UGI
W10-Adherence 0 UGI
W10-Direct 3 UGI
Writing 21.69 UGI

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.

Qwen/Qwen2.5-VL-32B-Instruct this lineage
AITRADER/Qwen2.5-VL-32B-Instruct-abliterated-mlx-8bit↓ 92

Metadata

License
apache-2.0
Languages
en
Tags
transformers safetensors qwen2_5_vl image-text-to-text multimodal abliterated uncensored mlx conversational en base_model:Qwen/Qwen2.5-VL-32B-Instruct base_model:quantized:Qwen/Qwen2.5-VL-32B-Instruct

Related

Other by AITRADER
Huihui-Qwen3.5-27B-Claude-4.6-Opus-abliterated-mlx-mxfp8 298 Jackrong-Qwen3.5-9B-Claude-Reasoning-abliterated-mxfp8-MLX 263 Huihui-Qwen3-Coder-Next-abliterated-mlx-8Bit 251 gemma-4-31B-Claude-Opus-Distill-abliterated-mlx-mxfp8 211 Qwen2.5-VL-32B-Instruct-abliterated-mlx-fp16 203
Same base model
huihui-ai/Qwen2.5-VL-32B-Instruct-abliterated 997 AITRADER/Qwen2.5-VL-32B-Instruct-abliterated-mlx-fp16 203 litmudoc/Qwen2.5-VL-32B-Instruct-abliterated-MLX-Q8 78
Total size
33.7 GB
Files
21
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-02-15 13:01
Source huggingface.co/AITRADER/Qwen2.5-VL-32B-Instruct-abliterated-mlx-8bit ↗ Base model Qwen/Qwen2.5-VL-32B-Instruct ↗

Files by quantization

Auxiliary files 21 files 33.7 GB
model-00004-of-00007.safetensors 5.00 GB 53dbec83 download
model-00005-of-00007.safetensors 5.00 GB 1d6ed179 download
model-00002-of-00007.safetensors 4.99 GB f3a3208a download
model-00001-of-00007.safetensors 4.98 GB ea668e1d download
model-00006-of-00007.safetensors 4.97 GB 645b3708 download
model-00003-of-00007.safetensors 4.97 GB b9052604 download
model-00007-of-00007.safetensors 3.81 GB 11ca5f95 download
tokenizer.json 10.9 MB 3fd16973 download
vocab.json 2.65 MB 4783fe10 download
model.safetensors.index.json 197 KB df24d7a9 download
chat_template.jinja 2.37 KB 28028c05 download
.gitattributes 1.53 KB 52373fe2 download
config.json 1.53 KB cf3e6f96 download
chat_template.json 1.03 KB 2ec1e359 download
README.md 766 B 03cb20e1 download
tokenizer_config.json 664 B 9fd0fb48 download
special_tokens_map.json 644 B 3a784031 download
added_tokens.json 629 B 06135f3c download
preprocessor_config.json 370 B c2d6bd6e download
generation_config.json 229 B 9b492590 download
configuration.json 76.0 B 4aef15d7 download

README current version from Hugging Face

license: apache-2.0
language:

  • en
    pipeline_tag: image-text-to-text
    tags:
  • multimodal
  • abliterated
  • uncensored
  • mlx
    library_name: transformers
    base_model:
  • Qwen/Qwen2.5-VL-32B-Instruct

AITRADER/Qwen2.5-VL-32B-Instruct-abliterated-mlx-8bit

This model was converted to MLX format from huihui-ai/Qwen2.5-VL-32B-Instruct-abliterated using mlx-vlm version 0.3.11.
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 AITRADER/Qwen2.5-VL-32B-Instruct-abliterated-mlx-8bit --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>

README history 1 version

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

  1. 2026-02-15Upload folder using huggingface_hubaf2c970766 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