← back to catalog · registered 2026-11-17 17:04

PrometheusEXE/DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic-i1-GGUF

PrometheusEXE Deepseek 4B GGUF 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/PrometheusEXE%2FDeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic-i1-GGUF"
Response includes
  • classification m3
  • files 18
  • author_summary 7 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
2
Likes
0
Model age
4d ago
created 2026-11-17
Downloads over time
Now 2 → from 0 ↑ 0%
0112 0 on Nov 17 2 on Nov 22 2 on Nov 19 Nov
Nov 17 → Nov 22 · 6 snapshots · spans 5 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 · 3 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 ko ja es ru
Quantizations
IQ3 IQ4 Q2_K Q3_K Q4 Q4_K Q5_K Q6_K
Tags
transformers gguf text-generation-inference unsloth qwen3_5 reasoning distillation deepseek sft mtp math stem

Related

Total size
36.5 GB
Files
18
Quantizations
10
Registered
2026-11-17 17:04
Last updated on HF
2026-11-17 16:41

Files by quantization

Q6_K 1 file 3.32 GB
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q6_K.gguf 3.32 GB ******** download
Q5_K 2 files 5.81 GB
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q5_K_M.gguf 2.94 GB ******** download
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q5_K_S.gguf 2.86 GB ******** download
Q4 2 files 5.09 GB
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q4_1.gguf 2.65 GB ******** download
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q4_0.gguf 2.44 GB ******** download
Q4_K 2 files 5.04 GB
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q4_K_M.gguf 2.59 GB ******** download
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q4_K_S.gguf 2.45 GB ******** download
IQ4 2 files 4.90 GB
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-IQ4_NL.gguf 2.49 GB ******** download
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-IQ4_XS.gguf 2.40 GB ******** download
Q3_K 3 files 6.45 GB
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q3_K_L.gguf 2.31 GB ******** download
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q3_K_M.gguf 2.16 GB ******** download
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q3_K_S.gguf 1.98 GB ******** download
IQ3 2 files 4.11 GB
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-IQ3_M.gguf 2.06 GB ******** download
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-IQ3_S.gguf 2.04 GB ******** download
Q2_K 1 file 1.82 GB
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.i1-Q2_K.gguf 1.82 GB ******** download
Drafts and companions 1 file 6.92 MB
DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic.imatrix.gguf 6.92 MB ******** download
Auxiliary files 2 files 8.91 KB
README.md 5.93 KB 57de0871 download
.gitattributes 2.99 KB 2f8caaec download

README current version from Hugging Face


base_model: xtools-at/DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic
language:

  • en
  • zh
  • ko
  • ja
  • es
  • ru
    library_name: transformers
    license: apache-2.0
    mradermacher:
    readme_rev: 1
    quantized_by: mradermacher
    tags:
  • text-generation-inference
  • transformers
  • unsloth
  • qwen3_5
  • reasoning
  • distillation
  • deepseek
  • sft
  • mtp
  • math
  • stem
  • heretic
  • uncensored
  • decensored
  • abliterated

About

weighted/imatrix quants of https://huggingface.co/xtools-at/DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic

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

static quants are available at https://huggingface.co/mradermacher/DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic-GGUF

This is a vision model - mmproj files (if any) will be in the static repository.

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-Q2_K 2.1 IQ3_XXS probably better
GGUF i1-Q3_K_S 2.2 IQ3_XS probably better
GGUF i1-IQ3_S 2.3 beats Q3_K*
GGUF i1-IQ3_M 2.3
GGUF i1-Q3_K_M 2.4 IQ3_S probably better
GGUF i1-Q3_K_L 2.6 IQ3_M probably better
GGUF i1-IQ4_XS 2.7
GGUF i1-Q4_0 2.7 fast, low quality
GGUF i1-Q4_K_S 2.7 optimal size/speed/quality
GGUF i1-IQ4_NL 2.8 prefer IQ4_XS
GGUF i1-Q4_K_M 2.9 fast, recommended
GGUF i1-Q4_1 2.9
GGUF i1-Q5_K_S 3.2
GGUF i1-Q5_K_M 3.3
GGUF i1-Q6_K 3.7 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.

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