Get the app
← back to catalog · registered 2026-08-22 13:56

Trilogix1/Hugstonized-qwen3.5-0.8B-abliterated-f32

Trilogix1 Qwen 800M GGUF 262K ctx
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/Trilogix1%2FHugstonized-qwen3.5-0.8B-abliterated-f32"
Response includes
  • classification m8
  • files 10
  • hub_downloads_all_time 1,130
  • author_summary 5 models
  • readme_text full
Try next
GET/models?author=Trilogix1
All 5 models by Trilogix1
 GET/models?family=qwen&method=m8
Sibling qwen models · same method
 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 · lifetime
1K
234 last 30d - stable
Likes
1
Model age
7mo ago
created 2026-03-05
Downloads over time
Now1.1K→from233↑392%
1875378871.2K233 on Mar 41.1K on Oct 111.1K on Oct 10MarAprMayJunJulAugSepOct
Mar 4 → Oct 11 · 71 snapshots · spans 221 days

Metadata

License
other
Quantizations
F16 IQ4 Q4_K Q5_K Q6_K Q8_0
Tags
gguf license:other endpoints_compatible region:us conversational

Related

Other by Trilogix1
Hugston-Huihui-Qwen3-4B-Thinking-2507-abliterated-f16 207 MiniMax-M2-abliterated-GGUF 24 DeepSeek-V3-Abliterated 22 GLM-5-abliterated 20
Total size
7.07 GB
Files
10
Quantizations
7
Registered
2026-08-22 13:56
Last updated on HF
2026-03-05 20:46
Source huggingface.co/Trilogix1/Hugstonized-qwen3.5-0.8B-abliterated-f32 ↗

Files by quantization

F16 1 file 1.41 GB
Hugstonized-qwen3.5-0.8B-abliterated-f32-f16.gguf 1.41 GB 4f634a47 download
Q8_0 1 file 774 MB
Hugstonized-qwen3.5-0.8B-abliterated-f32-Q8_0.gguf 774 MB 35d8dbc9 download
Q6_K 1 file 601 MB
Hugstonized-qwen3.5-0.8B-abliterated-f32-Q6_K.gguf 601 MB 1085f8a3 download
Q5_K 1 file 558 MB
Hugstonized-qwen3.5-0.8B-abliterated-f32-Q5_K_M.gguf 558 MB 913b4162 download
Q4_K 1 file 503 MB
Hugstonized-qwen3.5-0.8B-abliterated-f32-Q4_K_M.gguf 503 MB c66324a9 download
IQ4 1 file 480 MB
Hugstonized-qwen3.5-0.8B-abliterated-f32-IQ4_NL.gguf 480 MB d37b17da download
Auxiliary files 4 files 2.81 GB
Hugstonized-qwen3.5-0.8B-abliterated-f32.gguf 2.81 GB 0f957a26 download
.gitattributes 2.08 KB d1d98c68 download
README.md 690 B 47a5b0e6 download
LICENSE 0 B e69de29b download

README current version from Hugging Face

license: other
license_name: hugston
license_link: LICENSE

This model was converted and quantized with Quanta (A Hugston production). The weights were converted to F32 first.

The author of abliteration: https://huggingface.co/amkkk/qwen3.5-0.8b-abliterated-alllayers.

The model is for testing purposes only, used it at our own responsability and discretion.

Keep away from children.

Tested in abliteration queries in general,(seems to keep the accuracy, check example in screenshot).

This model can be used with HugstonOne.

Accuracy better at q6_k

image

README history 2 versions

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

  1. 2026-03-05Update README.mde17c517690 B
    Loading...
  2. 2026-03-05initial commit627cb0e67 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 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