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

huihui-ai/QwQ-32B-abliterated

huihui-ai Qwen 33B
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/huihui-ai%2FQwQ-32B-abliterated"
Response includes
  • classification m3
  • files 26
  • benchmarks 21 entries
  • hub_downloads_all_time 16,713
  • providers 1
  • author_summary 183 models
  • readme_text full
Try next
GET/models?author=huihui-ai
All 183 models by huihui-ai
 GET/models?family=qwen&method=m3
Sibling qwen models · same method
 GET/methods/m3
What "m3" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M3
Primary method

Layer-wise ablation

No other method signals detected in this model.
Confidence
HIGH
Why this label 3 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • author=huihui-ai (specializes in M3 layer-wise ablation)
  • is_gguf=0 (base model, not repackage)
  • 'abliterated' in name/tags
How we label → About M3 →
Refusal direction extracted via
Extraction technique

huihui-ai layer-band extraction

Fromhuihui-ai (2026) ↗
Confidence
HIGH
Why we say so
producer=huihui-ai (documented layer-band methodology in model cards)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
17K
148 last 30d - cooling
Likes
111
Descendants
36
in 17 direct forks
Model age
19mo ago
created 2025-03-07
Available via
1 provider
featherless-ai
Downloads over time
Now16.8K→from597↑2,706%
06.1K12.2K18.4K597 on Mar 5, 202516.8K on Oct 11Mar '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 17 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
huihui-ai/QwQ-32B-abliterated↓ 148 17 forks
DavidAU/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored↓ 35 13 forks
mradermacher/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-i1-GGUF↓ 900
DavidAU/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-gguf↓ 843
mradermacher/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-GGUF↓ 457
mlx-community/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-4bit↓ 86
alexgusevski/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-mlx-4Bit↓ 30
alexgusevski/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-mlx-fp16↓ 30
alexgusevski/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-mlx-8Bit↓ 26
Locosdeamor/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-310p-mlx-fp16↓ 26
alexgusevski/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-mlx-5Bit↓ 25
alexgusevski/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-mlx-3Bit↓ 24
alexgusevski/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-mlx-2Bit↓ 23
alexgusevski/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-mlx-6Bit↓ 19
RoadToNowhere/Qwen2.5-QwQ-35B-Eureka-Cubed-abliterated-uncensored-bnb-4bit↓ 15
DavidAU/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored↓ 31 6 forks
mradermacher/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored-i1-GGUF↓ 633
mradermacher/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored-GGUF↓ 422
cs2764/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored-mlx-6Bit-gs32↓ 42
DreamsOfControl/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored-mlx-4Bit↓ 30
Eric1227/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored_MLX_8bit↓ 29
Eric1227/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored_MLX_6bit↓ 23
bartowski/huihui-ai_QwQ-32B-abliterated-GGUF↓ 7,738
mradermacher/QwQ-32B-abliterated-i1-GGUF↓ 1,167
DevQuasar-9/huihui-ai.QwQ-32B-abliterated-GGUF↓ 444
mradermacher/QwQ-32B-abliterated-GGUF↓ 372
BenevolenceMessiah/QwQ-32B-abliterated-Q8_0-GGUF↓ 29
lactroiii/QwQ-32B-abliterated-i1-GGUF↓ 29
noneUsername/QwQ-32B-abliterated-AWQ-INT4↓ 26
awhiteside/QwQ-32B-abliterated-3bit↓ 19
KYUNGYONG/QwQ-32B-abliterated-4bit↓ 18
m-i/QwQ-32B-abliterated-mlx-8Bit↓ 17
ofer-tal/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored-3.8bpw-exl2↓ 16
KYUNGYONG/QwQ-32B-abliterated-3bit↓ 14
Uninformed/QwQ-32B-abliterated-exl2-5bpw-h8↓ 11
ofer-tal/Qwen2.5-QwQ-37B-Eureka-Triple-Cubed-abliterated-uncensored-4.0bpw-exl2↓ 10
noneUsername/QwQ-32B-abliterated-AWQ-INT4-float16↓ 6

Metadata

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

Related

Other by huihui-ai
Huihui-Qwen3.8-27B-abliterated-GGUF 2.2M Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF 351K Huihui-Qwen3.8-Flash-Next-abliterated-GGUF 331K Qwen2.5-VL-7B-Instruct-abliterated 301K Huihui-DeepSeek-V4-Flash-abliterated-ds4-GGUF 164K
Same base model
async0x42/QwQ-32B-abliterated-exl3_4.5bpw 40 TeeZee/QwQ-32B-abliterated-bpw4.0-h8-exl2 26 matatonic/QwQ-32B-abliterated-4.25bpw-exl2 21 thirdeyeai/QwQ-32B-uncensored 21 Iambackup/QwQ-32B-abliterated 19
Total size
61.0 GB
Files
26
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-03-12 04:47
Source huggingface.co/huihui-ai/QwQ-32B-abliterated ↗ Base model Qwen/QwQ-32B ↗

Files by quantization

Auxiliary files 26 files 61.0 GB
model-00001-of-00014.safetensors 4.56 GB ff30e2e5 download
model-00004-of-00014.safetensors 4.54 GB b0aeae8e download
model-00005-of-00014.safetensors 4.54 GB f2ffbaf5 download
model-00006-of-00014.safetensors 4.54 GB c9687954 download
model-00007-of-00014.safetensors 4.54 GB 4bc49168 download
model-00008-of-00014.safetensors 4.54 GB 723e78f6 download
model-00009-of-00014.safetensors 4.54 GB 100f7847 download
model-00010-of-00014.safetensors 4.54 GB fbb541d2 download
model-00011-of-00014.safetensors 4.54 GB aeb3564f download
model-00012-of-00014.safetensors 4.54 GB 4764d183 download
model-00013-of-00014.safetensors 4.54 GB f49621d4 download
model-00003-of-00014.safetensors 4.54 GB 364c4103 download
model-00002-of-00014.safetensors 4.54 GB fb67e595 download
model-00014-of-00014.safetensors 1.98 GB 8302838c download
tokenizer.json 10.9 MB aeb13307 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 8.16 KB da6dbc84 download
.gitattributes 1.53 KB 52373fe2 download
README.md 1.21 KB 1473fc5e download
added_tokens.json 735 B 6f359db5 download
config.json 690 B ea445f9c 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
    library_name: transformers

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

All versions of ollama from Q2_K to fp16 are supported.

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 4 versions

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

  1. 2025-03-12Update README.md24e340c1.2 KB
    Loading...
  2. 2025-03-08Update README.md3c427a81.2 KB
    Loading...
  3. 2025-03-07Upload 25 filesa7d3c5f1.1 KB
    Loading...
  4. 2025-03-07initial commit312e0be28 B
    Loading...

Discussions 7 threads

  1. 2025-04-03When doing roleplay, the model likes 'Counts to three' game/term. 当玩角色扮演时，模型很喜欢…open1 💬#7
    Loading...
  2. 2025-03-18Has the qwq32b-fp16 model on Ollama been updated?open2 💬#6
    Loading...
  3. 2025-03-15Great job on this! Thank you!!open5 💬#5
    Loading...
  4. 2025-03-12PRadd metadataopen1 💬#4
    Loading...
  5. 2025-03-10not as good as the previously released abliterated modelopen4 💬#3
    Loading...
  6. 2025-03-09There are plenty of "unsafe" prompts that don't get accepted with this model.open1 💬#2
    Loading...
  7. 2025-03-08Unable to generate adult contentopen5 💬#1
    Loading...

Continue on Hugging Face →

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