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

huihui-ai/DeepSeek-R1-Distill-Llama-8B-abliterated

huihui-ai Llama 8.0B
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-Llama-8B-abliterated"
Response includes
  • classification m3
  • files 13
  • benchmarks 5 entries
  • hub_downloads_all_time 19,017
  • 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=llama&method=m3
Sibling llama 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
19K
711 last 30d - cooling
Likes
79
Descendants
22
in 19 direct forks
Model age
20mo ago
created 2025-01-22
Available via
1 provider
featherless-ai
Downloads over time
Now19.1K→from178↑10,655%
07K14K21K178 on Jan 22, 202519.1K on Oct 11Jan '25Apr '25Jul '25Oct '25JanAprJulOct
Jan 22, 2025 → Oct 11 · 137 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 0.3111947074847581 OpenLLM-v2
IFEval instruct 0.44964028776978415 OpenLLM-v2
IFEval-Prompt 0.3068391866913124 OpenLLM-v2
MATH lvl 5 0 OpenLLM-v2
MMLU-Pro 0.20894281914893617 OpenLLM-v2

Genealogy 19 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-Llama-8B this lineage
huihui-ai/DeepSeek-R1-Distill-Llama-8B-abliterated↓ 711 19 forks
mergekit-community/Llama-3-ThinkRoleplay-DeepSeek-R1-Distill-8B-abliterated↓ 29 2 forks
mradermacher/Llama-3-ThinkRoleplay-DeepSeek-R1-Distill-8B-abliterated-i1-GGUF↓ 896
mradermacher/Llama-3-ThinkRoleplay-DeepSeek-R1-Distill-8B-abliterated-GGUF↓ 279
pharrow/DeepSeek-R1-Distill-Llama-8B-abliterated-SuperScaled↓ 19 1 fork
simeon-m/DeepSeek-R1-Distill-Llama-8B-abliterated-SuperScaled-Q4_K_M-GGUF↓ 18
cstr/DeepSeek-R1-Distill-Llama-8B-abliterated-Q4_K_M-GGUF↓ 746
tensorblock/DeepSeek-R1-Distill-Llama-8B-abliterated-GGUF↓ 287
PocketWeights/DeepSeek-R1-Llama-8B-Abliterated-GGUF↓ 205
maicog/DeepSeek-R1-Distill-Llama-8B-abliterated-Q4_K_S-GGUF↓ 174
AlfRjw/DeepSeek-R1-Distill-Llama-8B-abliterated-Q4-mlx↓ 45
kirihato/DeepSeek-R1-Distill-Llama-8B-abliterated-Q6_K-GGUF↓ 36
BernTheCreator/DeepSeek-R1-Distill-Llama-8B-abliterated-Q4_0-GGUF↓ 34
Astralyra/DeepSeek-R1-Distill-Llama-8B-abliterated-Q6_K-GGUF↓ 32
rottenorchid/DeepSeek-R1-Distill-Llama-8B-abliterated-Q4_K_M-GGUF↓ 32
AlfRjw/DeepSeek-R1-Distill-Llama-8B-abliterated-Q2-mlx↓ 31
kirihato/DeepSeek-R1-Distill-Llama-8B-abliterated-Q5_K_M-GGUF↓ 30
AlfRjw/DeepSeek-R1-Distill-Llama-8B-abliterated-Q3-mlx↓ 28
mergekit-community/Llama-3-LewdPlay-DeepSeek-R1-Distill-8B-abliterated↓ 24
bigwinboy/DeepSeek-R1-Distill-Llama-8B-abliterated-Q4_K_M-GGUF↓ 22
Human23/DeepSeek-R1-Distill-Llama-8B-abliterated-Q8_0-GGUF↓ 20
samwhra/DeepSeek-R1-Distill-Llama-8B-abliterated-Q4_K_M-GGUF↓ 15
numen-tech/DeepSeek-R1-Distill-Llama-8B-abliterated-w4a16g128asym

Metadata

Tags
transformers safetensors llama text-generation abliterated uncensored conversational base_model:deepseek-ai/DeepSeek-R1-Distill-Llama-8B base_model:finetune:deepseek-ai/DeepSeek-R1-Distill-Llama-8B 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
UnfilteredAI/DAN-L3-R1-8B 4K AIOpsInSpace/DeepSeek-R1-Distill-Llama-8B-Abliterated-Patched 341 stepenZEN/DeepSeek-R1-Distill-Llama-8B-Abliterated 161 DevQuasar-3/DevQuasar-R1-Uncensored-Llama-8B 36 Sangsang/R1-8B-thinksafe-DeepSeek-8B-unfiltered-32-pm-3ep 20
Total size
15.0 GB
Files
13
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-02-16 08:38
Source huggingface.co/huihui-ai/DeepSeek-R1-Distill-Llama-8B-abliterated ↗ Base model deepseek-ai/DeepSeek-R1-Distill-Llama-8B ↗

Files by quantization

Auxiliary files 13 files 15.0 GB
model-00002-of-00004.safetensors 4.66 GB 75bbe8ad download
model-00001-of-00004.safetensors 4.63 GB 5a90e908 download
model-00003-of-00004.safetensors 4.58 GB 498d24ba download
model-00004-of-00004.safetensors 1.09 GB eb0cadcd download
tokenizer.json 16.4 MB d9191504 download
tokenizer_config.json 53.7 KB fc5a3326 download
model.safetensors.index.json 23.7 KB c4b1c21a download
.gitattributes 1.53 KB 52373fe2 download
README.md 1.16 KB 88f71d95 download
LICENSE 1.04 KB d42fae90 download
config.json 860 B 02a43a06 download
special_tokens_map.json 508 B 6caea2ce download
generation_config.json 190 B 5d9190fb download

README current version from Hugging Face

base_model:

  • deepseek-ai/DeepSeek-R1-Distill-Llama-8B
    library_name: transformers
    tags:
  • abliterated
  • uncensored

huihui-ai/DeepSeek-R1-Distill-Llama-8B-abliterated

This is an uncensored version of deepseek-ai/DeepSeek-R1-Distill-Llama-8B 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.

If "<think>" does not appear or refuses to respond, you can first provide an example to guide, and then ask your question.
For instance:

  How many 'r' characters are there in the word "strawberry"?

Use with ollama

You can use huihui_ai/deepseek-r1-abliterated directly

ollama run huihui_ai/deepseek-r1-abliterated:8b

Donation

Your donation helps us continue our further development and improvement, a cup of coffee can do it.
  • bitcoin:
  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge

README history 12 versions

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

  1. 2025-02-16Update README.mdae88f141.2 KB
    Loading...
  2. 2025-02-16Update README.md173d3641.2 KB
    Loading...
  3. 2025-02-16Update README.md9052e9f1.2 KB
    Loading...
  4. 2025-02-16Update README.mdfbef0801.2 KB
    Loading...
  5. 2025-02-16Update README.md6a02d8f1.2 KB
    Loading...
  6. 2025-02-16Update README.md4566f6e1.1 KB
    Loading...
  7. 2025-02-09Update README.md31e10bb970 B
    Loading...
  8. 2025-02-09Update README.mda833b40952 B
    Loading...
  9. 2025-02-09Update README.mda7c7bf5942 B
    Loading...
  10. 2025-01-25Update README.mde8f16a5785 B
    Loading...
  11. 2025-01-23Upload 12 files596ec76599 B
    Loading...
  12. 2025-01-22initial commit516ca0328 B
    Loading...

Discussions 4 threads

  1. 2026-05-07PRFix chat_template crash when assistant message omits the `content` keyopen1 💬#4
    Loading...
  2. 2025-02-06零度解說發影片了open2 💬#3
    Loading...
  3. 2025-01-30This model till censor the contentopen5 💬#2
    Loading...
  4. 2025-01-24Ollama availabilityopen2 💬#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