← back to catalog · registered 2026-09-12 10:55

ailexleon/Dark-Nexus-27B-v3.0-uncensored-heretic-mlx-6Bit

ailexleon 27B multimodal
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
57
Likes
0
Model age
2d ago
created 2026-09-12
Downloads over time
Now57from0↑0%
02142630 on Sep 1257 on Sep 14Sep
Sep 12 → Sep 14 · 3 snapshots · spans 2 days

Metadata

Tags
mlx safetensors qwen3_5 creative writing roleplay conversational character-rp storytelling image-text-to-text en base_model:llmfan46/Dark-Nexus-27B-v3.0-uncensored-heretic base_model:quantized:llmfan46/Dark-Nexus-27B-v3.0-uncensored-heretic

Related

Total size
21.2 GB
Files
14
Quantizations
1
Registered
2026-09-12 10:55
Last updated on HF
2026-09-12 11:03

Files by quantization

Auxiliary files 14 files 21.2 GB
model-00001-of-00005.safetensors 4.99 GB 6701de02 download
model-00002-of-00005.safetensors 4.98 GB 1078c263 download
model-00004-of-00005.safetensors 4.97 GB 72d03ef3 download
model-00003-of-00005.safetensors 4.97 GB 81e2b7eb download
model-00005-of-00005.safetensors 1.31 GB 3effd32b download
tokenizer.json 19.1 MB 639e352c download
model.safetensors.index.json 213 KB 8b98b99f download
chat_template.jinja 7.68 KB 57fcf921 download
config.json 4.86 KB bb7dc93d download
.gitattributes 1.53 KB 52373fe2 download
tokenizer_config.json 1.24 KB 9de16b5b download
processor_config.json 989 B 823b73aa download
README.md 395 B e703e8e5 download
generation_config.json 226 B 7c8bef13 download

README current version from Hugging Face


language: en
library_name: mlx
pipeline_tag: image-text-to-text
tags:

  • mlx

ailexleon/Dark-Nexus-27B-v3.0-uncensored-heretic-mlx-6Bit

Use with mlx

pip install -U mlx-vlm
python -m mlx_vlm.generate --model ailexleon/Dark-Nexus-27B-v3.0-uncensored-heretic-mlx-6Bit --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>

README history 3 versions

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

  1. 2026-09-12Update README.mdd6246f1717 B
    Loading...
  2. 2026-09-12Update README.md1b17641675 B
    Loading...
  3. 2026-09-12Upload folder using huggingface_hubdb0650f395 B
    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.