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

SicariusSicariiStuff/Wingless_Imp_8B_Abliterated

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/SicariusSicariiStuff%2FWingless_Imp_8B_Abliterated"
Response includes
  • classification m1
  • files 9
  • benchmarks 11 entries
  • hub_downloads_all_time 1,163
  • author_summary 75 models
  • readme_text full
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)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · lifetime
1K
108 last 30d - cooling
Likes
4
Descendants
3
in 3 direct forks
Model age
8mo ago
created 2026-01-27
Downloads over time
Now1.2K→from8↑14,650%
04328651.3K8 on Jan 281.2K on Oct 12JanMarMayJulSep
Jan 28 → Oct 12 · 77 snapshots · spans 257 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 1.8 UGI
Hazardous 1.2 UGI
Natural Intelligence 11 UGI
Political lean -1.8% UGI
Sensitive-Info 16.46 UGI
SocPol 1.9 UGI
UGI 33.47 UGI
Willingness (10) 6.8 UGI
W10-Adherence 5.5 UGI
W10-Direct 8 UGI
Writing 23.51 UGI

Genealogy 3 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.

Metadata

License
llama3.1
Languages
en
Tags
safetensors llama en base_model:SicariusSicariiStuff/Wingless_Imp_8B base_model:finetune:SicariusSicariiStuff/Wingless_Imp_8B license:llama3.1 region:us

Related

Total size
15.0 GB
Files
9
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-04-24 11:25

Files by quantization

Auxiliary files 9 files 15.0 GB
model.safetensors 15.0 GB 02eed6e0 download
tokenizer.json 16.4 MB 65ff5472 download
tokenizer_config.json 49.4 KB 3beeacc8 download
README.md 2.74 KB 329bb2e9 download
.gitattributes 1.59 KB 90183944 download
config.json 868 B 0e49f0db download
chat_template.jinja 348 B 7402f8c0 download
special_tokens_map.json 325 B b43be966 download
generation_config.json 155 B e1ce4973 download

README current version from Hugging Face


license: llama3.1
language:

base_model:

  • SicariusSicariiStuff/Wingless_Imp_8B

Wingless_Imp_8B
Abliterated

Developed by: SicariusSicariiStuff

Wingless_Imp_8B_Abliterated is an abliterated variant of SicariusSicariiStuff/Wingless_Imp_8B with surgical removal of refusal mechanisms. This model maintains the full capabilities of the original, while eliminating safety guardrails through orthogonalization techniques.

KL divergence

<0.02

Refusals

~10%

What is KL divergence?

Think about it as a way to measure the variance between the original model "World Model," vs the abliterated one; the lower the KL divergence, the closer the "World Model" of the two models to each other.

If the original model thinks making pineapple pizza is a crime against humanity (it is), then the abliterated model will still hold to this belief, but if asked how to make one (probably after giving you a disclaimer about what an abomination that is), it would still tell you how. In other words, most of the knowledge, quirks, and capabilities are preserved.


Technical Specs

  • Base Model: Wingless_Imp_8B
  • Parameters: 8B
  • Context Length: 128K tokens
  • Architecture: Llama (decoder-only transformer)
  • Precision: fp16
  • Method: Orthogonalization-based abliteration
  • License: Llama 3.1 Community License

Methodology

  1. Identifies refusal direction vectors in activation space
  2. Orthogonalizes weights to inhibit activation along these directions
  3. Preserves (mostly) all other model behaviors and knowledge

Model Details

  • Intended use: General Tasks, Roleplay.

  • Censorship level: Very Low

  • X / 10 (10 completely uncensored)

UGI score:

pending evals


Citation Information

@llm{Wingless_Imp_8B_Abliterated,
  author = {SicariusSicariiStuff},
  title = {Wingless_Imp_8B_Abliterated},
  year = {2026},
  publisher = {Hugging Face},
  url = {https://huggingface.co/SicariusSicariiStuff/Wingless_Imp_8B_Abliterated}
}

Other stuff

README history 2 versions

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

  1. 2026-04-24Update README.md0fc0ae32.7 KB
    Loading...
  2. 2026-01-27Create README.md9f0cbc22.7 KB
    Loading...

Discussions 2 threads

  1. 2026-04-30Abliterationopen1 💬#2
    Loading...
  2. 2026-04-24Error during quantizationclosed4 💬#1
    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.

Open in Abliteration