Get the app
← back to catalog · registered 2026-09-19 11:56

maxylium/Qwythos-9B-v2-Heretic-Abliterated-GGUF

maxylium 9B GGUF second-order 
curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/maxylium%2FQwythos-9B-v2-Heretic-Abliterated-GGUF"
Response includes
  • classification m3
  • files 6
  • author_summary 1 models
  • readme_text full
Try next
GET/methods/m3
What "m3" means · taxonomy
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
M1 Direct removal →
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)
How we label → About M3 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · 30-day
0
Likes
0
Model age
today
created 2026-09-19

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

usermma/Qwythos-9B-v2-Heretic-Abliterated↓ 66 this lineage
maxylium/Qwythos-9B-v2-Heretic-Abliterated-GGUF

Metadata

License
other
Quantizations
Q2_K Q4_K Q5_K Q8_0
Tags
gguf llama-cpp quantized text-generation base_model:usermma/Qwythos-9B-v2-Heretic-Abliterated base_model:quantized:usermma/Qwythos-9B-v2-Heretic-Abliterated license:other endpoints_compatible region:us conversational
Total size
23.7 GB
Files
6
Quantizations
5
Registered
2026-09-19 11:56
Last updated on HF
2026-09-19 11:53
Source huggingface.co/maxylium/Qwythos-9B-v2-Heretic-Abliterated-GGUF ↗ Base model usermma/Qwythos-9B-v2-Heretic-Abliterated →

Files by quantization

Q8_0 1 file 8.87 GB
Qwythos-9B-v2-Heretic-Abliterated-Q8_0.gguf 8.87 GB 5f30382b download
Q5_K 1 file 6.02 GB
Qwythos-9B-v2-Heretic-Abliterated-Q5_K_M.gguf 6.02 GB 3e3f43d9 download
Q4_K 1 file 5.24 GB
Qwythos-9B-v2-Heretic-Abliterated-Q4_K_M.gguf 5.24 GB 9e7efea4 download
Q2_K 1 file 3.56 GB
Qwythos-9B-v2-Heretic-Abliterated-Q2_K.gguf 3.56 GB b61fc454 download
Auxiliary files 2 files 2.35 KB
.gitattributes 1.80 KB 4a3a5a53 download
README.md 562 B f12f5cb4 download

README current version from Hugging Face

base_model: usermma/Qwythos-9B-v2-Heretic-Abliterated
tags:

  • gguf
  • llama-cpp
  • quantized
  • text-generation
    pipeline_tag: text-generation
    license: other

Qwythos-9B-v2-Heretic-Abliterated-GGUF

quantized GGUF weights for usermma/Qwythos-9B-v2-Heretic-Abliterated.

available quants:

  • Qwythos-9B-v2-Heretic-Abliterated-Q4_K_M.gguf
  • Qwythos-9B-v2-Heretic-Abliterated-Q8_0.gguf
  • Qwythos-9B-v2-Heretic-Abliterated-Q2_K.gguf
  • Qwythos-9B-v2-Heretic-Abliterated-Q5_K_M.gguf
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.

See supported apps Browse the catalog