← back to catalog · registered 2026-09-12 17:55

AIconjured/Qwen3.8-27B-TURBO-Fable-ColdFusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-NVFP4-Q8-Q3-GGUF

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 · lifetime
345
Likes
5
Model age
1d ago
created 2026-09-12
Downloads over time
Now345from0↑0%
01272533800 on Sep 12345 on Sep 14Sep
Sep 12 → Sep 14 · 3 snapshots · spans 2 days

Metadata

Tags
gguf 27B Qwen3.8 MTP Text Vision Ollama llama.cpp Uncensored NVFP4 blackwell Text Generation

Related

Total size
0 B
Files
1
Quantizations
1
Registered
2026-09-12 17:55
Last updated on HF
2026-09-13 01:43

Files by quantization

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

README

The author did not publish a README.

README history 6 versions

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

  1. 2026-09-13Update README.md268c60b9.2 KB
    Loading...
  2. 2026-09-12Update README.mda2a5aca7.3 KB
    Loading...
  3. 2026-09-12Update README.md48c50a67.3 KB
    Loading...
  4. 2026-09-12Update README.mda04a5cc7.3 KB
    Loading...
  5. 2026-09-12Update README.md3e1ed5e7 KB
    Loading...
  6. 2026-09-12Upload folder using huggingface_hub056b0127.1 KB
    Loading...

Discussions 1 thread

  1. 2026-09-12num_ctx 229376 fits comfortably on 32GB Vram for 100% GPUopen2 💬#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 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.