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

huihui-ai/Huihui-Qwen3.5-2B-abliterated

huihui-ai Qwen 2.3B 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/huihui-ai%2FHuihui-Qwen3.5-2B-abliterated"
Response includes
  • classification m3
  • files 13
  • benchmarks 11 entries
  • hub_downloads_all_time 21,420
  • author_summary 184 models
  • readme_text full
Try next
GET/models?author=huihui-ai
All 184 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
21K
2K last 30d - cooling
HF trending score
1
mild recent interest
Likes
51
Descendants
12
in 9 direct forks
Model age
7mo ago
created 2026-03-03
Downloads over time
Now21.9K→from0↑0%
08K16.1K24.1K0 on Mar 321.9K on Oct 11MarAprMayJunJulAugSepOct
Mar 3 → Oct 11 · 73 snapshots · spans 222 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 0.3 UGI
Hazardous 0.6 UGI
Natural Intelligence 9.03 UGI
Political lean -14.3% UGI
Sensitive-Info 2.6 UGI
SocPol 0 UGI
UGI 5.9 UGI
Willingness (10) 1.2 UGI
W10-Adherence 1.5 UGI
W10-Direct 1 UGI
Writing 19.27 UGI

Genealogy 9 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/Qwen3.5-2B this lineage
huihui-ai/Huihui-Qwen3.5-2B-abliterated↓ 1,530 9 forks
nbeerbower/Huihui-Qwen3.5-2B-abliterated-Athanorlite-ORPO↓ 28 2 forks
mradermacher/Huihui-Qwen3.5-2B-abliterated-Athanorlite-ORPO-i1-GGUF↓ 2,038
mradermacher/Huihui-Qwen3.5-2B-abliterated-Athanorlite-ORPO-GGUF↓ 923
brishen/iron-huihui-qwen3.5-2b-abliterated-npu2 1 fork
Atomic-Germ/HuiHui-Qwen3.5-Abliterated-2B-NPU2↓ 33
mradermacher/Huihui-Qwen3.5-2B-abliterated-GGUF↓ 9,522
mradermacher/Huihui-Qwen3.5-2B-abliterated-i1-GGUF↓ 1,509
interimlabs/InterimLabs-Huihui-Qwen3.5-2B-abliterated-GGUF-Q4_K_M↓ 487
noTrabajo/Huihui-Qwen3.5-2B-abliterated-Q4_K_M-GGUF↓ 234
interimlabs/InterimLabs-Huihui-Qwen3.5-2B-abliterated-MLX-4bit↓ 153
Giniiki/Huihui-Qwen3.5-2B-abliterated-mlx-4bit↓ 148
RaiseRuntimeError/Huihui-Qwen3.5-2B-abliterated-q4f16_1-MLC↓ 116

Metadata

License
apache-2.0
Tags
transformers safetensors qwen3_5 image-text-to-text abliterated uncensored conversational base_model:Qwen/Qwen3.5-2B base_model:finetune:Qwen/Qwen3.5-2B license:apache-2.0 endpoints_compatible region:us

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
Nic22Of2003/Qwen3.5-2B-Uncensored-HauhauCS-Aggressive-IQ4_NL 984 prithivMLmods/Gliese-Qwen3.5-2B-Abliterated-Caption 744 arzaan789/qwen3.5-2b-uncensored_ 707 DreamFast/Qwen3.5-2B-Uncensored-HauhauCS-Aggressive-Safetensor-Benchmark 634 arzaan789/qwen3.5-2b-uncensored 623
Total size
4.24 GB
Files
13
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-03-03 08:16
Source huggingface.co/huihui-ai/Huihui-Qwen3.5-2B-abliterated ↗ Base model Qwen/Qwen3.5-2B ↗

Files by quantization

Auxiliary files 13 files 4.26 GB
model.safetensors-00001-of-00001.safetensors 4.24 GB 568f9529 download
tokenizer.json 12.2 MB 5f9e4d49 download
vocab.json 6.41 MB 0aa0ce06 download
merges.txt 3.20 MB a494e019 download
model.safetensors.index.json 62.9 KB 69a41758 download
tokenizer_config.json 16.3 KB fae3ce99 download
LICENSE 11.3 KB f938136e download
chat_template.jinja 7.57 KB 0ef09f21 download
config.json 2.84 KB d30a15be download
README.md 2.41 KB 0846faba download
.gitattributes 1.53 KB 52373fe2 download
preprocessor_config.json 390 B 2ea84a43 download
video_preprocessor_config.json 385 B 3ba673a5 download

README current version from Hugging Face

library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3.5-2B/blob/main/LICENSE
pipeline_tag: image-text-to-text
base_model:

  • Qwen/Qwen3.5-2B
    tags:
  • abliterated
  • uncensored

huihui-ai/Huihui-Qwen3.5-2B-abliterated

This is an uncensored version of Qwen/Qwen3.5-2B 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.

ollama

Please use the latest version of ollama v0.17.5

You can use huihui_ai/qwen3.5-abliterated:2b directly, 

ollama run huihui_ai/qwen3.5-abliterated:2b

Usage Warnings

  • Risk of Sensitive or Controversial Outputs: This model’s safety filtering has been significantly reduced, potentially generating sensitive, controversial, or inappropriate content. Users should exercise caution and rigorously review generated outputs.

  • Not Suitable for All Audiences: Due to limited content filtering, the model’s outputs may be inappropriate for public settings, underage users, or applications requiring high security.

  • Legal and Ethical Responsibilities: Users must ensure their usage complies with local laws and ethical standards. Generated content may carry legal or ethical risks, and users are solely responsible for any consequences.

  • Research and Experimental Use: It is recommended to use this model for research, testing, or controlled environments, avoiding direct use in production or public-facing commercial applications.

  • Monitoring and Review Recommendations: Users are strongly advised to monitor model outputs in real-time and conduct manual reviews when necessary to prevent the dissemination of inappropriate content.

  • No Default Safety Guarantees: Unlike standard models, this model has not undergone rigorous safety optimization. huihui.ai bears no responsibility for any consequences arising from its use.

Donation

Your donation helps us continue our further development and improvement, a cup of coffee can do it.
  • bitcoin:
  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge
  • Support our work on Ko-fi!

README history 2 versions

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

  1. 2026-03-03Update README.mdb2e291a2.4 KB
    Loading...
  2. 2026-03-03Create README.md0d978e72.1 KB
    Loading...

Discussions 1 thread

  1. 2026-04-06PRUpload options.jsonclosed1 💬#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