← back to catalog · registered 2026-09-18 23:56

OS-Software/llm-jp-4-32b-a3b-thinking-uncensored-heretic

OS-Software 32B MoE
curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/OS-Software%2Fllm-jp-4-32b-a3b-thinking-uncensored-heretic"
Response includes
  • classification m3
  • files 1
  • author_summary 18 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M3
Primary method

Layer-wise ablation

Applied on top of direct removal inherited from the base model.
Confidence
HIGH
Inherited from base model
Why this label 2 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • 'heretic' in model name (Heretic-produced)
  • Heretic uses layer-wise optimization (M3) with underlying direction removal (M1)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
Downloads · 30-day
0
Likes
0
Model age
today
created 2026-09-18
Downloads over time
Now0from0↑0%
00110 on Sep 180 on Sep 19Sep
Sep 18 → Sep 19 · 2 snapshots · spans 1 day

Metadata

Tags
transformers safetensors qwen3_moe text-generation heretic uncensored decensored abliterated conversational en ja license:apache-2.0

Related

Total size
0 B
Files
1
Quantizations
1
Registered
2026-09-18 23:56
Last updated on HF
2026-09-19 00:14

Files by quantization

Auxiliary files 1 file 1.48 KB
.gitattributes 1.48 KB a6344aac download

README

The author did not publish a README.

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 app" button that hands off directly to a local runtime of your choice - Infrahuman, LM Studio, or Ollama. No API keys, no subscription, no prompt leakage.