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

sulpikar2/Qwen3.5-9B-Uncensored-cyber-v3

sulpikar2 Qwen 9.4B multimodal second-order
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/sulpikar2%2FQwen3.5-9B-Uncensored-cyber-v3"
Response includes
  • classification m1
  • files 12
  • hub_downloads_all_time 478
  • author_summary 6 models
  • readme_text full
Try next
GET/models?author=sulpikar2
All 6 models by sulpikar2
 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
478
47 last 30d - cooling
Likes
1
Descendants
1
in 1 direct fork
Model age
5mo ago
created 2026-04-29
Downloads over time
Now491→from195↑152%
180294407521195 on Apr 29491 on Oct 11491 on Oct 9AprMayJunJulAugSepOct
Apr 29 → Oct 11 · 63 snapshots · spans 165 days

Genealogy 1 direct fork

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

sulpikar2/Qwen3.5-9B-Uncensored-cyber↓ 20 this lineage
sulpikar2/Qwen3.5-9B-Uncensored-cyber-v3↓ 47 1 fork
mradermacher/Qwen3.5-9B-Uncensored-cyber-v3-GGUF↓ 1,116

Metadata

License
apache-2.0
Languages
en
Tags
transformers safetensors qwen3_5 image-text-to-text text-generation-inference unsloth heretic uncensored decensored abliterated reproducible conversational

Related

Other by sulpikar2
Qwen3.6-27B-Ultimate-Uncensored-heretic 35 Huihui-Qwen3.5-9B-Claude-4.6-Opus-abliterated-heretic 33 agent-Qwen3.5-9B-Claude-4.6-Opus-abliterated-heretic 30 Qwen3.5-9B-Uncensored-cyber 20 Qwen3.5-9B-Uncensored-cyber-v2 9
Same base model
mradermacher/Qwen3.5-9B-Uncensored-cyber-GGUF 649 sulpikar2/Qwen3.5-9B-Uncensored-cyber-v2 9
Total size
17.5 GB
Files
12
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-04-29 21:24
Source huggingface.co/sulpikar2/Qwen3.5-9B-Uncensored-cyber-v3 ↗ Base model sulpikar2/Qwen3.5-9B-Uncensored-cyber →

Files by quantization

Auxiliary files 12 files 17.5 GB
model-00002-of-00004.safetensors 4.65 GB 38735877 download
model-00003-of-00004.safetensors 4.61 GB 25d53285 download
model-00001-of-00004.safetensors 4.60 GB f0f7ebe9 download
model-00004-of-00004.safetensors 3.66 GB 3572a7db download
tokenizer.json 19.1 MB 639e352c download
model.safetensors.index.json 67.6 KB 778b7bbb download
chat_template.jinja 7.57 KB a585dec8 download
config.json 2.87 KB d6c8fa4f download
README.md 1.75 KB 3b3b8b2f download
.gitattributes 1.53 KB 52373fe2 download
tokenizer_config.json 1.14 KB e187ba1e download
generation_config.json 141 B 35a874f7 download

README current version from Hugging Face

base_model: sulpikar2/Qwen3.5-9B-Uncensored-cyber
tags:

  • text-generation-inference
  • transformers
  • unsloth
  • qwen3_5
  • heretic
  • uncensored
  • decensored
  • abliterated
  • reproducible
    license: apache-2.0
    language:
  • en

This is a decensored version of sulpikar2/Qwen3.5-9B-Uncensored-cyber-v2, made using Heretic v1.2.0

[!TIP]
This model is reproducible!

See the README in the reproduce directory for more information.

Abliteration parameters

Parameter Value
direction_index per layer
attn.o_proj.max_weight 1.10
attn.o_proj.max_weight_position 23.47
attn.o_proj.min_weight 0.44
attn.o_proj.min_weight_distance 14.96
mlp.down_proj.max_weight 0.93
mlp.down_proj.max_weight_position 30.61
mlp.down_proj.min_weight 0.18
mlp.down_proj.min_weight_distance 15.94

Performance

Metric This model Original model (sulpikar2/Qwen3.5-9B-Uncensored-cyber-v2)
KL divergence 0.0576 0 (by definition)
Refusals 8/100 23/100

Uploaded finetuned model

  • Developed by: sulpikar2
  • License: apache-2.0
  • Finetuned from model : sulpikar2/Qwen3.5-9B-Uncensored-cyber

This qwen3_5 model was trained 2x faster with Unsloth and Huggingface's TRL library.

README history 2 versions

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

  1. 2026-04-29Upload README.md with huggingface_hub9292bc71.7 KB
    Loading...
  2. 2026-04-29Upload Qwen3_5ForConditionalGeneration968edc95.1 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