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

matatonic/QwQ-32B-abliterated-4.25bpw-exl2

matatonic Qwen 32B
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/matatonic%2FQwQ-32B-abliterated-4.25bpw-exl2"
Response includes
  • classification m1
  • files 15
  • benchmarks 21 entries
  • hub_downloads_all_time 48
  • author_summary 16 models
  • readme_text full
Try next
GET/models?author=matatonic
All 16 models by matatonic
 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
48
21 last 30d - stable
Likes
4
Model age
19mo ago
created 2025-03-07
Downloads over time
Now55→from21↑162%
0469313921 on Mar 5, 202555 on Oct 11127 on Oct 1, 2025Mar '25Jun '25Sep '25Dec '25MarJunSep
Mar 5, 2025 → Oct 11 · 123 snapshots · spans 585 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
Arena-Battles 24359 LM-Arena
LM Arena Elo 1333.1043906834145 LM-Arena
Arena-Elo-Lower 1328.5976095286157 LM-Arena
Arena-Elo-Upper 1337.6111718382135 LM-Arena
Arena-Rank 50 LM-Arena
BBH average 0.3017252561940589 OpenLLM-v2
IFEval instruct 0.47002398081534774 OpenLLM-v2
IFEval-Prompt 0.32532347504621073 OpenLLM-v2
MATH lvl 5 0.1608761329305136 OpenLLM-v2
MMLU-Pro 0.11959773936170212 OpenLLM-v2
Entertainment 1.4 UGI
Hazardous 4.1 UGI
Natural Intelligence 24.06 UGI
Political lean -18.6% UGI
Sensitive-Info 27.21 UGI
SocPol 3.2 UGI
UGI 27.3 UGI
Willingness (10) 2.8 UGI
W10-Adherence 2.5 UGI
W10-Direct 3 UGI
Writing 34.87 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/QwQ-32B this lineage
matatonic/QwQ-32B-abliterated-4.25bpw-exl2↓ 21

Metadata

License
apache-2.0
Languages
en
Tags
safetensors qwen2 chat abliterated uncensored text-generation conversational en base_model:Qwen/QwQ-32B base_model:quantized:Qwen/QwQ-32B license:apache-2.0 exl2

Related

Other by matatonic
GLM-4-32B-0414-abliterated-Q6_K-GGUF 69 GLM-4-32B-0414-abliterated-Q5_K_S-GGUF 46 GLM-4-32B-0414-abliterated-Q8_0-GGUF 44 DeepSeek-R1-Distill-Llama-70B-abliterated-6.5bpw-h8-exl2 33 Qwen2.5-72B-Instruct-abliterated-v2-6.5bpw-h8-exl2 26
Same base model
huihui-ai/QwQ-32B-abliterated 148 async0x42/QwQ-32B-abliterated-exl3_4.5bpw 40 TeeZee/QwQ-32B-abliterated-bpw4.0-h8-exl2 26 thirdeyeai/QwQ-32B-uncensored 21 Iambackup/QwQ-32B-abliterated 19
Total size
17.5 GB
Files
15
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-03-07 22:41
Source huggingface.co/matatonic/QwQ-32B-abliterated-4.25bpw-exl2 ↗ Base model Qwen/QwQ-32B ↗

Files by quantization

Auxiliary files 15 files 17.5 GB
output-00001-of-00002.safetensors 10.00 GB 93f3a18e download
output-00002-of-00002.safetensors 7.47 GB adfd55ff download
tokenizer.json 10.9 MB aeb13307 download
measurement.json 3.41 MB 3413187c download
vocab.json 2.65 MB 4783fe10 download
merges.txt 1.59 MB 31349551 download
model.safetensors.index.json 62.5 KB d19f75d8 download
LICENSE 11.1 KB 352f53db download
tokenizer_config.json 8.15 KB 7d4bc82f download
.gitattributes 1.53 KB 52373fe2 download
README.md 1.13 KB 95ffca2e download
config.json 979 B 882f7a08 download
added_tokens.json 735 B 6f359db5 download
special_tokens_map.json 644 B 3a784031 download
generation_config.json 257 B bcd663ca download

README current version from Hugging Face

license: apache-2.0
license_link: https://huggingface.co/huihui-ai/QwQ-32B-abliterated/blob/main/LICENSE
language:

  • en
    pipeline_tag: text-generation
    base_model: Qwen/QwQ-32B
    tags:
  • chat
  • abliterated
  • uncensored

huihui-ai/QwQ-32B-abliterated

This is an uncensored version of Qwen/QwQ-32B created with abliteration (see remove-refusals-with-transformers to know more about it).
This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.

Use with ollama

You can use huihui_ai/qwq-abliterated directly

ollama run huihui_ai/qwq-abliterated

Donation

If you like it, please click 'like' and follow us for more updates.
You can follow x.com/support_huihui to get the latest model information from huihui.ai.

Your donation helps us continue our further development and improvement, a cup of coffee can do it.
  • bitcoin:
  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge

README history 1 version

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

  1. 2025-03-07initialb41d8631.1 KB
    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