← back to catalog · registered 2026-09-17 17:56

marcorez8/GLM-5.3-Flash-Abliterated-Heretic-V2-LoRA-v5-UD-IQ4_XS

marcorez8 Glm 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 · 30-day
0
Likes
0
Model age
today
created 2026-09-17

Metadata

Tags
gguf region:us

Related

Total size
146 GB
Files
7
Quantizations
2
Registered
2026-09-17 17:56
Last updated on HF
2026-09-17 17:56

Files by quantization

F16 1 file 1.05 GB
mmproj-F16.gguf 1.05 GB 96ccc182 download
Auxiliary files 6 files 146 GB
GLM-5.3-Flash-Abliterated-Heretic-V2-UD-IQ4_XS-00002-of-00005.gguf 46.6 GB f6d89fd6 download
GLM-5.3-Flash-Abliterated-Heretic-V2-UD-IQ4_XS-00003-of-00005.gguf 46.2 GB eaa7626c download
GLM-5.3-Flash-Abliterated-Heretic-V2-UD-IQ4_XS-00004-of-00005.gguf 46.1 GB ccc2e051 download
GLM-5.3-Flash-Abliterated-Heretic-V2-UD-IQ4_XS-00005-of-00005.gguf 7.20 GB 66ebf9ec download
GLM-5.3-Flash-Abliterated-Heretic-V2-UD-IQ4_XS-00001-of-00005.gguf 8.99 MB eec97673 download
.gitattributes 2.52 KB bb2a894d 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.