← back to catalog · registered 2026-11-18 15:04

MaximKrupkin/Qwen3.6-35B-A3B-uncensored-heretic-Q5_K_M-MACaQ-R-GGUF

MaximKrupkin 35B GGUF MoE 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/MaximKrupkin%2FQwen3.6-35B-A3B-uncensored-heretic-Q5_K_M-MACaQ-R-GGUF"
Response includes
  • classification m3
  • files 3
  • benchmarks 11 entries
  • author_summary 2 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
277
Likes
1
Model age
3d ago
created 2026-11-18
Downloads over time
Now 277 → from 0 ↑ 0%
0102203305 0 on Nov 18 277 on Nov 22 Nov
Nov 18 → Nov 22 · 5 snapshots · spans 4 days

Benchmarks

Benchmark Score Source
Entertainment 1.9 UGI
Hazardous 2.4 UGI
Natural Intelligence 28.42 UGI
Political lean -12.5% UGI
Sensitive-Info 18.88 UGI
SocPol 1.5 UGI
UGI 44.25 UGI
Willingness (10) 9.5 UGI
W10-Adherence 9 UGI
W10-Direct 10 UGI
Writing 37.6 UGI

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.

Metadata

License
apache-2.0
Quantizations
Q5_K
Tags
gguf macaq macaq-r qwen3_5_moe mixture-of-experts heretic uncensored abliterated text-generation arxiv:2611.10649 base_model:llmfan46/Qwen3.6-35B-A3B-uncensored-heretic base_model:quantized:llmfan46/Qwen3.6-35B-A3B-uncensored-heretic

Related

Total size
23.1 GB
Files
3
Quantizations
2
Registered
2026-11-18 15:04
Last updated on HF
2026-11-18 14:56

Files by quantization

Q5_K 1 file 23.1 GB
Qwen3.6-35B-A3B-uncensored-heretic-Q5_K_M-MACaQ-R.gguf 23.1 GB d4f9aabd download
Auxiliary files 2 files 4.89 KB
README.md 3.32 KB 826896ac download
.gitattributes 1.57 KB 30c38d2b download

README current version from Hugging Face


license: apache-2.0
base_model:

  • llmfan46/Qwen3.6-35B-A3B-uncensored-heretic
    base_model_relation: quantized
    library_name: gguf
    pipeline_tag: text-generation
    tags:
  • gguf
  • macaq
  • macaq-r
  • qwen3_5_moe
  • mixture-of-experts
  • heretic
  • uncensored
  • abliterated

Qwen3.6-35B-A3B-uncensored-heretic — Q5_K_M MACaQ-R (GGUF)

A MACaQ-R refined version of the Q5_K_M quant of
llmfan46/Qwen3.6-35B-A3B-uncensored-heretic.

MACaQ-R (Huawei CSL, paper) does not
quantize from scratch: it refines the weight codes of an existing native K-quant MoE GGUF (Q4_K / Q5_K / Q6_K) against a
full-precision teacher, keeping block scales, tensor formats and the original activation basis. All other tensors are
copied unchanged. The result loads in standard llama.cpp.

Both the HF reference checkpoint (--hf) and the teacher (--teacher) were built from the heretic weights, not from
stock Qwen3.6, so the refinement targets the abliterated model's own distribution.

File

File Size SHA-256
Qwen3.6-35B-A3B-uncensored-heretic-Q5_K_M-MACaQ-R.gguf 24,764,032,768 bytes d4f9aabd08f8cec3580c3d53d60b28a450e86d6f7c5404014371eff2a3d2a15c

How it was made

  • Source quant: Qwen3.6-35B-A3B-uncensored-heretic-Q5_K_M.gguf from
    llmfan46/Qwen3.6-35B-A3B-uncensored-heretic-GGUF
  • HF reference: llmfan46/Qwen3.6-35B-A3B-uncensored-heretic (bf16 safetensors)
  • Teacher: f16 GGUF converted from the same heretic safetensors (convert_hf_to_gguf.py --outtype f16 --no-mtp)
  • Calibration: examples/make_moe_calibration.py --profile diverse --seed 617029 --count 6000
  • Refine command (MACaQ llamacpp/macaq.py, defaults otherwise: context 512, 256 chunks, 64 iterations):
python llamacpp/macaq.py refine \
  --model Qwen3.6-35B-A3B-uncensored-heretic-Q5_K_M.gguf \
  --hf models/heretic-hf --teacher models/heretic-f16.gguf \
  --calibration data/calibration.txt --gpus 0,1 --device cuda \
  --gpu-layers 28 --ubatch 512 --threads 48 \
  --output Qwen3.6-35B-A3B-uncensored-heretic-Q5_K_M-MACaQ-R.gguf --work-dir work/refine
  • Refined: 370 K-quant weight tensors (120 expert tensors × 256 experts).
  • Hardware: 2× RTX 3090.

Usage

llama-server -m Qwen3.6-35B-A3B-uncensored-heretic-Q5_K_M-MACaQ-R.gguf -ngl 99

Smoke test (llama.cpp, 2× RTX 3090, -ngl 99): loads in ~18 s, ~139 tokens/s generation.

Evaluation status

Quality has not been measured yet for this file (no KL-divergence / perplexity numbers versus the f16 teacher, and
no check of refusal behaviour). For reference, Huawei reports for stock Qwen3.6-35B-A3B at the same 22.36 GB size a
KL-divergence to FP16 of 0.013367 for MACaQ-R Q4_K_XL vs 0.015731 for Unsloth UD-Q4_K_XL; that number is not a claim
about this file.

License

Apache-2.0, following the licenses of
Qwen/Qwen3.6-35B-A3B,
llmfan46/Qwen3.6-35B-A3B-uncensored-heretic
and MACaQ.

This is an uncensored (abliterated) model. You are responsible for how you use it.

README history 1 version

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

  1. 2026-11-18 Add model card (Apache-2.0) 1d2657b 3.3 KB
    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