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

huihui-ai/DeepSeek-R1-Distill-Qwen-14B-abliterated

huihui-ai Qwen 15B
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/huihui-ai%2FDeepSeek-R1-Distill-Qwen-14B-abliterated"
Response includes
  • classification m3
  • files 15
  • benchmarks 16 entries
  • hub_downloads_all_time 3,194
  • providers 1
  • author_summary 183 models
  • readme_text full
Try next
GET/models?author=huihui-ai
All 183 models by huihui-ai
 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

No other method signals detected in this model.
Confidence
HIGH
Why this label 3 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • author=huihui-ai (specializes in M3 layer-wise ablation)
  • is_gguf=0 (base model, not repackage)
  • 'abliterated' in name/tags
How we label → About M3 →
Refusal direction extracted via
Extraction technique

huihui-ai layer-band extraction

Fromhuihui-ai (2026) ↗
Confidence
HIGH
Why we say so
producer=huihui-ai (documented layer-band methodology in model cards)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
3K
229 last 30d - cooling
Likes
34
Descendants
6
in 6 direct forks
Model age
20mo ago
created 2025-01-22
Available via
1 provider
featherless-ai
Downloads over time
Now3.3K→from135↑2,326%
01.2K2.4K3.6K135 on Jan 22, 20253.3K on Oct 11Jan '25Apr '25Jul '25Oct '25JanAprJulOct
Jan 22, 2025 → Oct 11 · 129 snapshots · spans 627 days

Benchmarks

Portrait before abliteration
Benchmarks of the base model as it stood before the refusal-removal operation. Compare with the numbers above to see what the operation cost.
Benchmark Score Source
BBH average 0.5542871967829809 OpenLLM-v2
IFEval instruct 0.5047961630695443 OpenLLM-v2
IFEval-Prompt 0.37153419593345655 OpenLLM-v2
MATH lvl 5 0.4086102719033233 OpenLLM-v2
MMLU-Pro 0.4666722074468085 OpenLLM-v2
Entertainment 1.5 UGI
Hazardous 2.4 UGI
Natural Intelligence 16.39 UGI
Political lean -27.4% UGI
Sensitive-Info 15.62 UGI
SocPol 1 UGI
UGI 21.25 UGI
Willingness (10) 3.2 UGI
W10-Adherence 0.5 UGI
W10-Direct 6 UGI
Writing 26.29 UGI

Genealogy 6 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.

deepseek-ai/DeepSeek-R1-Distill-Qwen-14B this lineage
huihui-ai/DeepSeek-R1-Distill-Qwen-14B-abliterated↓ 229 6 forks
mradermacher/DeepSeek-R1-Distill-Qwen-14B-abliterated-i1-GGUF↓ 2,415
mradermacher/DeepSeek-R1-Distill-Qwen-14B-abliterated-GGUF↓ 688
CuckmeisterFuller/DeepSeek-R1-Distill-Qwen-14B-abliterated-Q4-mlx↓ 370
tensorblock/DeepSeek-R1-Distill-Qwen-14B-abliterated-GGUF↓ 149
JASONWORLD/DeepSeek-R1-Distill-Qwen-14B-abliterated-Q4_K_M-GGUF↓ 25
numen-tech/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2-GPTQ-Int4

Metadata

Tags
transformers safetensors qwen2 text-generation abliterated uncensored conversational base_model:deepseek-ai/DeepSeek-R1-Distill-Qwen-14B base_model:finetune:deepseek-ai/DeepSeek-R1-Distill-Qwen-14B text-generation-inference endpoints_compatible region:us

Related

Other by huihui-ai
Huihui-Qwen3.8-27B-abliterated-GGUF 2.2M Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF 351K Huihui-Qwen3.8-Flash-Next-abliterated-GGUF 331K Qwen2.5-VL-7B-Instruct-abliterated 301K Huihui-DeepSeek-V4-Flash-abliterated-ds4-GGUF 164K
Same base model
QuantFactory/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2-GGUF 4K huihui-ai/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2 1K TenTour/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2-GGUF 64 thirdeyeai/DeepSeek-R1-Distill-Qwen-14B-uncensored 46 Greytechai/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2 26
Total size
27.5 GB
Files
15
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-01-23 13:31
Source huggingface.co/huihui-ai/DeepSeek-R1-Distill-Qwen-14B-abliterated ↗ Base model deepseek-ai/DeepSeek-R1-Distill-Qwen-14B ↗

Files by quantization

Auxiliary files 15 files 27.5 GB
model-00001-of-00006.safetensors 4.64 GB 9e5980e5 download
model-00003-of-00006.safetensors 4.61 GB 7a62b8ac download
model-00004-of-00006.safetensors 4.61 GB efd7b782 download
model-00005-of-00006.safetensors 4.61 GB dce933dd download
model-00002-of-00006.safetensors 4.61 GB cdb2c2a3 download
model-00006-of-00006.safetensors 4.41 GB 11b46f3c download
tokenizer.json 10.9 MB e20ddafc download
model.safetensors.index.json 46.9 KB 915d5cc3 download
tokenizer_config.json 6.79 KB f8c31060 download
.gitattributes 1.53 KB 52373fe2 download
LICENSE 1.04 KB d42fae90 download
README.md 818 B 80811fbd download
config.json 691 B b2dd8e73 download
special_tokens_map.json 508 B 6caea2ce download
generation_config.json 190 B 829e028b download

README current version from Hugging Face

base_model:

  • deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
    library_name: transformers
    tags:
  • abliterated
  • uncensored

huihui-ai/DeepSeek-R1-Distill-Qwen-14B-abliterated

This is an uncensored version of deepseek-ai/DeepSeek-R1-Distill-Qwen-14B created with abliteration (see remove-refusals-with-transformers to know more about it).
This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.

Important Note There's a new version available, please try using the new version huihui-ai/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2.

README history 3 versions

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

  1. 2025-01-23Update README.md65f0318818 B
    Loading...
  2. 2025-01-22Upload 14 files2162ea7599 B
    Loading...
  3. 2025-01-22initial commitebb243e28 B
    Loading...

Discussions 2 threads

  1. 2025-12-17PRUpdate README.mdopen1 💬#2
    Loading...
  2. 2025-01-23the abliterated is not very thoroughopen9 💬#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