static quants of https://huggingface.co/OS-Software/ThinkingCap-Qwen3.8-27B-Uncensored-Heretic
mradermacher/ThinkingCap-Qwen3.8-27B-Uncensored-Heretic-GGUF
curl -H "Authorization: Bearer $ABL_KEY" \
"https://abliteration.org/api/v1/models/mradermacher%2FThinkingCap-Qwen3.8-27B-Uncensored-Heretic-GGUF" Response includes
- classification m3
- files 6
- author_summary 3254 models
- readme_text full
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
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 2026-09-24
Metadata
Related
Files by quantization
Q4_K 1 file 14.5 GB
| ThinkingCap-Qwen3.8-27B-Uncensored-Heretic.Q4_K_S.gguf | 14.5 GB | b3bc4428 | download |
Q2_K 1 file 9.98 GB
| ThinkingCap-Qwen3.8-27B-Uncensored-Heretic.Q2_K.gguf | 9.98 GB | e98b5806 | download |
F16 1 file 885 MB
| ThinkingCap-Qwen3.8-27B-Uncensored-Heretic.mmproj-f16.gguf | 885 MB | 21f35cd1 | download |
Q8_0 1 file 600 MB
| ThinkingCap-Qwen3.8-27B-Uncensored-Heretic.mmproj-Q8_0.gguf | 600 MB | 69ad8827 | download |