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

mradermacher/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored-GGUF

mradermacher Qwen 32B GGUF second-order 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/mradermacher%2FQwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored-GGUF"
Response includes
  • classification m3
  • files 15
  • hub_downloads_all_time 3,136
  • author_summary 3324 models
  • readme_text full
Try next
GET/models?author=mradermacher
All 3324 models by mradermacher
 GET/models?family=qwen&method=m3
Sibling qwen models · same method
 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 · lifetime
3K
803 last 30d - stable
Likes
0
Model age
7mo ago
created 2026-02-28

Training datasets

Corpora the author lists in the model card. Datasets tracked in our /datasets catalog carry a category badge linking to the workflow stage. Others open on Hugging Face.

TeichAI/polaris-alpha-1000x
Downloads over time
Now3.4K→from0↑0%
01.2K2.5K3.7K0 on Feb 253.4K on Oct 11FebAprJunAugOct
Feb 25 → Oct 11 · 72 snapshots · spans 228 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.

DavidAU/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored this lineage
mradermacher/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored-GGUF↓ 803

Variants by this author 2 formats · 3K 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 zh
Quantizations
IQ4 Q2_K Q3_K Q4_K Q5_K Q6_K Q8_0
Tags
transformers gguf unsloth heretic uncensored abliterated fine tune creative creative writing fiction writing plot generation sub-plot generation

Related

Other by mradermacher
Ornith-1.5-9B-uncensored-GGUF 1.5M Qwen3-VL-8B-Instruct-abliterated-GGUF 667K Qwen3.8-Flash-Next-Uncensored-i1-GGUF 153K gemma-4-26B-A4B-it-ultra-uncensored-heretic-i1-GGUF 138K Qwen3.8-9B-heretic-uncensored-GGUF 106K
Same base model
mradermacher/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored-i1-GGUF 2K nightmedia/Qwen3-32B-VL-Instruct-Polaris-GLM-Heretic-Uncensored-qx86-hi-mlx 212 nightmedia/Qwen3-32B-VL-Instruct-Polaris-GLM-Heretic-Uncensored-qx64-hi-mlx 204 yybl/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored-Q4_K_M-GGUF 163 yybl/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored-Q5_K_M-GGUF 159
Total size
209 GB
Files
15
Quantizations
9
Registered
2026-08-22 13:56
Last updated on HF
2026-02-28 15:11
Source huggingface.co/mradermacher/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored-GGUF ↗ Base model DavidAU/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored ↗

Files by quantization

Q8_0 2 files 33.1 GB
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.Q8_0.gguf 32.4 GB 70a0b559 download
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.mmproj-Q8_0.gguf 737 MB df302499 download
Q6_K 1 file 25.0 GB
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.Q6_K.gguf 25.0 GB 6a09e902 download
Q5_K 2 files 42.7 GB
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.Q5_K_M.gguf 21.6 GB 08219672 download
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.Q5_K_S.gguf 21.1 GB 63ca194a download
Q4_K 2 files 35.9 GB
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.Q4_K_M.gguf 18.4 GB 44b53d26 download
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.Q4_K_S.gguf 17.5 GB b798166e download
IQ4 1 file 16.6 GB
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.IQ4_XS.gguf 16.6 GB 1ea69950 download
Q3_K 3 files 44.4 GB
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.Q3_K_L.gguf 16.1 GB 4314c3d8 download
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.Q3_K_M.gguf 14.9 GB 511880ed download
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.Q3_K_S.gguf 13.4 GB c0de3deb download
Q2_K 1 file 11.5 GB
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.Q2_K.gguf 11.5 GB 238cadab download
F16 1 file 1.11 GB
Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored.mmproj-f16.gguf 1.11 GB 7f56f08d download
Auxiliary files 2 files 7.95 KB
README.md 5.17 KB 7f63eece download
.gitattributes 2.78 KB f0c64b2b download

README current version from Hugging Face

base_model: DavidAU/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored
datasets:

  • TeichAI/polaris-alpha-1000x
    language:
  • en
  • zh
    library_name: transformers
    license: apache-2.0
    mradermacher:
    readme_rev: 1
    quantized_by: mradermacher
    tags:
  • unsloth
  • heretic
  • uncensored
  • abliterated
  • fine tune
  • creative
  • creative writing
  • fiction writing
  • plot generation
  • sub-plot generation
  • fiction writing
  • story generation
  • scene continue
  • storytelling
  • fiction story
  • science fiction
  • romance
  • all genres
  • story
  • writing
  • vivid prosing
  • vivid writing
  • fiction
  • roleplaying
  • bfloat16
  • swearing
  • rp
  • horror
  • r rated
  • x rated
  • all use cases

About

static quants of https://huggingface.co/DavidAU/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored

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

weighted/imatrix quants are available at https://huggingface.co/mradermacher/Qwen3-32B-VL-Instruct-Polaris-HI16-Heretic-Uncensored-i1-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 mmproj-Q8_0 0.9 multi-modal supplement
GGUF mmproj-f16 1.3 multi-modal supplement
GGUF Q2_K 12.4
GGUF Q3_K_S 14.5
GGUF Q3_K_M 16.1 lower quality
GGUF Q3_K_L 17.4
GGUF IQ4_XS 18.0
GGUF Q4_K_S 18.9 fast, recommended
GGUF Q4_K_M 19.9 fast, recommended
GGUF Q5_K_S 22.7
GGUF Q5_K_M 23.3
GGUF Q6_K 27.0 very good quality
GGUF Q8_0 34.9 fast, best quality

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.

README history 4 versions

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

  1. 2026-02-28auto-patch README.md8dbbb765.2 KB
    Loading...
  2. 2026-02-28auto-patch README.mda5f3cff5.3 KB
    Loading...
  3. 2026-02-28auto-patch README.md2f2de0e5.1 KB
    Loading...
  4. 2026-02-28uploaded from nico105197fa404 B
    Loading...

Discussions 1 thread

  1. 2026-05-03Corrupted model?open1 💬#1
    Loading...

Continue on Hugging Face →

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