Get the app
← back to catalog · registered 2026-11-06 19:02

Dhdjxkxdnsbzj/lfm3-1.7b-abliterated-gguf

Dhdjxkxdnsbzj 1.7B GGUF
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model. 

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/Dhdjxkxdnsbzj%2Flfm3-1.7b-abliterated-gguf"
Response includes
  • classification m8
  • files 5
  • author_summary 1 models
  • readme_text full
Try next
GET/methods/m8
What "m8" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

Applied on top of direct removal inherited from the base model.
Confidence
MEDIUM
Inherited from base model
M1 Direct removal →
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=1
  • assume M1 (base ablation) + M8 (GGUF quant) - default when producer unknown
How we label → About M8 →
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · 30-day
0
Likes
0
Model age
today
created 2026-11-06

Metadata

Languages
en
Quantizations
F16 Q4_K
Tags
gguf lfm3 abliteration research en endpoints_compatible region:us conversational
Total size
4.25 GB
Files
5
Quantizations
3
Registered
2026-11-06 19:02
Last updated on HF
2026-11-06 18:30
Source huggingface.co/Dhdjxkxdnsbzj/lfm3-1.7b-abliterated-gguf ↗

Files by quantization

F16 1 file 3.23 GB
lfm3-1.7b-abliterated.f16.gguf 3.23 GB dec51a23 download
Q4_K 1 file 1.02 GB
lfm3-1.7b-abliterated.Q4_K_M.gguf 1.02 GB 7b5bf54a download
Auxiliary files 3 files 2.70 KB
.gitattributes 1.62 KB 89b712c4 download
README.md 803 B d165e6d1 download
Modelfile.lfm3-abliterated 309 B 9ea18d18 download

README current version from Hugging Face

language:

  • en
    tags:
  • lfm3
  • abliteration
  • gguf
  • research

LFM3-1.7B abliterated - f16 GGUF

Refusal direction (layer 13, 64 harmful vs 64 harmless prompts) orthogonalised out of all
49 residual-stream writers. Measured on 40 held-out harmful prompts:

base abliterated
refusal rate 100% (40/40) 22.5% (9/40)
perplexity 10.60 11.44

f16 is a floating-point format, not a quantised one.

huggingface-cli download Dhdjxkxdnsbzj/lfm3-1.7b-abliterated-gguf \
    --repo-type model --local-dir ./abl
ollama create lfm3-abliterated -f ./abl/Modelfile.lfm3-abliterated

The chat template defaults enable_thinking=false (LFM3 otherwise emits an open thinking
and loops). For tool calling use llama-server --jinja or Ollama's /v1 endpoint.

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 Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration