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

stepenZEN/DeepSeek-R1-Distill-Llama-8B-Abliterated

stepenZEN 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/stepenZEN%2FDeepSeek-R1-Distill-Llama-8B-Abliterated"
Response includes
  • classification m1
  • files 12
  • benchmarks 5 entries
  • hub_downloads_all_time 7,374
  • providers 1
  • author_summary 7 models
  • readme_text full
Try next
GET/models?author=stepenZEN
All 7 models by stepenZEN
 GET/models?family=llama&method=m1
Sibling llama models · same method
 GET/methods/m1
What "m1" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
How we label → About M1 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
7K
161 last 30d - cooling
Likes
10
Descendants
10
in 10 direct forks
Model age
20mo ago
created 2025-01-20
Available via
1 provider
featherless-ai
Downloads over time
Now7.4K→from0↑0%
02.7K5.4K8.2K0 on Jan 15, 20257.4K on Oct 12Jan '25Apr '25Jul '25Oct '25JanAprJulOct
Jan 15, 2025 → Oct 12 · 131 snapshots · spans 635 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 10 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
stepenZEN/DeepSeek-R1-Distill-Llama-8B-Abliterated↓ 161 10 forks
mradermacher/DeepSeek-R1-Distill-Llama-8B-Abliterated-GGUF↓ 4,750
mradermacher/DeepSeek-R1-Distill-Llama-8B-Abliterated-i1-GGUF↓ 3,737
ufydy937/DeepSeek-R1-Distill-Llama-8B-Abliterated-GGUF↓ 1,034
data-pioneer-0826/DeepSeek-R1-Distill-Llama-8B-Abliterated-GGUF↓ 434
matrixportalx/DeepSeek-R1-Distill-Llama-8B-Abliterated-GGUF↓ 306
ibrahimkettaneh/DeepSeek-R1-Distill-Llama-8B-Abliterated-Q4_K_M-GGUF↓ 114
matrixportalx/DeepSeek-R1-Distill-Llama-8B-Abliterated-Q4_K_M-GGUF↓ 82
CuckmeisterFuller/DeepSeek-R1-Distill-Llama-8B-Abliterated-Q4-mlx↓ 65
Iambackup/DeepSeek-R1-Distill-Llama-8B-Abliterated-GGUF↓ 55
matrixportalx/DeepSeek-R1-Distill-Llama-8B-Abliterated-Q4_0-GGUF↓ 54

Metadata

Languages
en
Tags
transformers safetensors llama text-generation conversational en 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 stepenZEN
DeepSeek-R1-Distill-Qwen-1.5B-Abliterated-dpo 49 Qwen2.5-0.5B-abliterated 29 DeepSeek-R1-Distill-Qwen-32B-abliterated 24 Qwen2.5-3B-Instruct-Abliterated 21 DeepSeek-R1-Distill-Llama-70B-Abliterated 17
Same base model
UnfilteredAI/DAN-L3-R1-8B 4K huihui-ai/DeepSeek-R1-Distill-Llama-8B-abliterated 711 AIOpsInSpace/DeepSeek-R1-Distill-Llama-8B-Abliterated-Patched 341 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
12
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-01-20 18:23
Source huggingface.co/stepenZEN/DeepSeek-R1-Distill-Llama-8B-Abliterated ↗ Base model deepseek-ai/DeepSeek-R1-Distill-Llama-8B ↗

Files by quantization

Auxiliary files 12 files 15.0 GB
model-00002-of-00004.safetensors 4.66 GB 1bf4944f download
model-00001-of-00004.safetensors 4.63 GB e82f835d download
model-00003-of-00004.safetensors 4.58 GB 43d1a0d8 download
model-00004-of-00004.safetensors 1.09 GB d3039cc6 download
tokenizer.json 16.4 MB 26c881aa download
tokenizer_config.json 51.7 KB e02acbac download
model.safetensors.index.json 23.4 KB 0fd8120f download
.gitattributes 1.53 KB 52373fe2 download
config.json 886 B b65cd60f download
special_tokens_map.json 371 B d53f816e download
README.md 153 B bdeb9bb9 download
generation_config.json 121 B 61c68659 download

README current version from Hugging Face

language:

  • en
    base_model:
  • deepseek-ai/DeepSeek-R1-Distill-Llama-8B
    library_name: transformers

DeepSeek-R1-Distill-Llama-8B-Abliterated

README history 1 version

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

  1. 2025-01-20Create README.md94c4fb2153 B
    Loading...

Discussions 2 threads

  1. 2025-01-27abliteration ineffective?open2 💬#2
    Loading...
  2. 2025-01-2470B Abliteratedopen2 💬#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