← back to catalog · registered 2026-11-20 00:04

mradermacher/Qwen3.5-0.8B-Heretic-Uncensored-i1-GGUF

mradermacher 800M GGUF multimodal second-order
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/mradermacher%2FQwen3.5-0.8B-Heretic-Uncensored-i1-GGUF"
Response includes
  • classification m3
  • files 27
  • author_summary 3602 models
  • readme_text full
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
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
3K
Likes
2
Model age
2d ago
created 2026-11-19
Downloads over time
Now 3.3K → from 0 ↑ 0%
01.2K2.5K3.7K 0 on Nov 20 3.3K on Nov 22 Nov
Nov 20 → Nov 22 · 3 snapshots · spans 2 days

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.

Variants by this author 2 formats · 5K downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

License
apache-2.0
Languages
en
Quantizations
IQ1 IQ2 IQ3 IQ4 Q2_K Q3_K Q4 Q4_K Q5_K Q6_K
Tags
transformers gguf qwen qwen3.5 qwen3_5 multimodal vision-language conversational safetensors heretic uncensored decensored

Related

Total size
10.2 GB
Files
27
Quantizations
12
Registered
2026-11-20 00:04
Last updated on HF
2026-11-20 21:32

Files by quantization

Q6_K 1 file 601 MB
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q6_K.gguf 601 MB 2108efa2 download
Q5_K 2 files 1.06 GB
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q5_K_M.gguf 551 MB b07bb791 download
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q5_K_S.gguf 538 MB 36b97384 download
Q4 2 files 987 MB
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q4_1.gguf 508 MB c3cbb5a7 download
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q4_0.gguf 479 MB 1d02af50 download
Q4_K 2 files 987 MB
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q4_K_M.gguf 505 MB 542c9d81 download
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q4_K_S.gguf 482 MB 146dc1c0 download
IQ4 2 files 973 MB
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ4_NL.gguf 492 MB b722fda0 download
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ4_XS.gguf 481 MB 65f651a9 download
Q3_K 3 files 1.30 GB
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q3_K_L.gguf 469 MB 8e332be6 download
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q3_K_M.gguf 445 MB cfdb7f51 download
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q3_K_S.gguf 415 MB ace3bd20 download
IQ3 4 files 1.63 GB
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ3_M.gguf 433 MB 81a2b2e9 download
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ3_S.gguf 430 MB ccb6f637 download
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ3_XS.gguf 423 MB 1cbd3ecb download
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ3_XXS.gguf 380 MB 07c6ea4d download
Q2_K 2 files 798 MB
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q2_K.gguf 403 MB f7a1e202 download
Qwen3.5-0.8B-Heretic-Uncensored.i1-Q2_K_S.gguf 395 MB 0a25ed55 download
IQ2 4 files 1.36 GB
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ2_M.gguf 360 MB 32a3fb6f download
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ2_S.gguf 350 MB d5ff5e52 download
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ2_XS.gguf 347 MB c27e8fdc download
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ2_XXS.gguf 336 MB f9a65af3 download
IQ1 2 files 637 MB
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ1_M.gguf 323 MB d260620b download
Qwen3.5-0.8B-Heretic-Uncensored.i1-IQ1_S.gguf 315 MB 94394ef1 download
Drafts and companions 1 file 2.17 MB
Qwen3.5-0.8B-Heretic-Uncensored.imatrix.gguf 2.17 MB 2e85e370 download
Auxiliary files 2 files 10.3 KB
README.md 6.84 KB 3be6e175 download
.gitattributes 3.50 KB 466b0824 download

README current version from Hugging Face


base_model: Peymandc/Qwen3.5-0.8B-Heretic-Uncensored
language:

  • en
    library_name: transformers
    license: apache-2.0
    license_link: https://huggingface.co/Qwen/Qwen3.5-0.8B/blob/main/LICENSE
    mradermacher:
    readme_rev: 1
    quantized_by: mradermacher
    tags:
  • qwen
  • qwen3.5
  • qwen3_5
  • multimodal
  • vision-language
  • conversational
  • transformers
  • safetensors
  • heretic
  • uncensored
  • decensored
  • abliterated
  • reproducible

About

weighted/imatrix quants of https://huggingface.co/Peymandc/Qwen3.5-0.8B-Heretic-Uncensored

For a convenient overview and download list, visit our model page for this model.

static quants are available at https://huggingface.co/mradermacher/Qwen3.5-0.8B-Heretic-Uncensored-GGUF

Usage

If you are unsure how to use GGUF files, refer to one of TheBloke's
READMEs
for
more details, including on how to concatenate multi-part files.

Provided Quants

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

Link Type Size/GB Notes
GGUF imatrix 0.1 imatrix file (for creating your own quants)
GGUF i1-IQ1_S 0.4 for the desperate
GGUF i1-IQ1_M 0.4 mostly desperate
GGUF i1-IQ2_XXS 0.5
GGUF i1-IQ2_XS 0.5
GGUF i1-IQ2_S 0.5
GGUF i1-IQ2_M 0.5
GGUF i1-IQ3_XXS 0.5 lower quality
GGUF i1-Q2_K_S 0.5 very low quality
GGUF i1-Q2_K 0.5 IQ3_XXS probably better
GGUF i1-Q3_K_S 0.5 IQ3_XS probably better
GGUF i1-IQ3_XS 0.5
GGUF i1-IQ3_S 0.6 beats Q3_K*
GGUF i1-IQ3_M 0.6
GGUF i1-Q3_K_M 0.6 IQ3_S probably better
GGUF i1-Q3_K_L 0.6 IQ3_M probably better
GGUF i1-Q4_0 0.6 fast, low quality
GGUF i1-IQ4_XS 0.6
GGUF i1-Q4_K_S 0.6 optimal size/speed/quality
GGUF i1-IQ4_NL 0.6 prefer IQ4_XS
GGUF i1-Q4_K_M 0.6 fast, recommended
GGUF i1-Q4_1 0.6
GGUF i1-Q5_K_S 0.7
GGUF i1-Q5_K_M 0.7
GGUF i1-Q6_K 0.7 practically like static Q6_K

Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):

image.png

And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9

FAQ / Model Request

See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.

Thanks

I thank my company, nethype GmbH, for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.

README history 4 versions

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

  1. 2026-11-20 auto-patch README.md 2857684 7 KB
    Loading...
  2. 2026-11-20 auto-patch README.md be1d87c 6.8 KB
    Loading...
  3. 2026-11-19 auto-patch README.md 1265aaa 2.6 KB
    Loading...
  4. 2026-11-19 uploaded from back 9537d9e 486 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.

Open in Abliteration