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

Jellon/Qwen2.5-72B-Instruct-abliterated-exl2-3bpw

Jellon Qwen 72B 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/Jellon%2FQwen2.5-72B-Instruct-abliterated-exl2-3bpw"
Response includes
  • classification m1
  • files 16
  • benchmarks 5 entries
  • hub_downloads_all_time 189
  • author_summary 4 models
  • readme_text full
Try next
GET/models?author=Jellon
All 4 models by Jellon
 GET/models?family=qwen&method=m1
Sibling qwen 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
189
14 last 30d - cooling
Likes
1
Model age
24mo ago
created 2024-10-21
Downloads over time
Now196→from1↑19,500%
0721442161 on Oct 16, 2024196 on Oct 12196 on Oct 11Oct '24Feb '25Jun '25Oct '25FebJunOct
Oct 16, 2024 → Oct 12 · 144 snapshots · spans 726 days

Benchmarks

Benchmark Score Source
BBH average 0.6423660656375665 OpenLLM-v2
IFEval instruct 0.7613908872901679 OpenLLM-v2
IFEval-Prompt 0.6691312384473198 OpenLLM-v2
MATH lvl 5 0.46148036253776437 OpenLLM-v2
MMLU-Pro 0.5871841755319149 OpenLLM-v2

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.

zetasepic/Qwen2.5-72B-Instruct-abliterated↓ 80 this lineage
Jellon/Qwen2.5-72B-Instruct-abliterated-exl2-3bpw↓ 14

Metadata

License
other
Languages
en
Tags
safetensors qwen2 chat text-generation conversational en base_model:zetasepic/Qwen2.5-72B-Instruct-abliterated base_model:quantized:zetasepic/Qwen2.5-72B-Instruct-abliterated license:other 3-bit exl2 region:us

Related

Other by Jellon
Qwen2.5-14B-Instruct-abliterated-v2-6bpw 21 Llama-3.1-Nemotron-70B-Instruct-HF-abliterated-exl2-3bpw 17 Qwen2.5-32B-Instruct-abliterated-exl2-4bpw 15
Same base model
featherless-ai-quants/zetasepic-Qwen2.5-72B-Instruct-abliterated-GGUF 304 cgifbribcgfbi/Qwen2.5-72B-Instruct-abliterated-GPTQ-INT8 82 cgifbribcgfbi/Qwen2.5-72B-Instruct-abliterated-chem-claude-5-comp3-sort-pat 21 Zenabius/Qwen2.5-72B-Instruct-abliterated-exl2-6.5bpw 11 cgifbribcgfbi/Qwen2.5-72B-Instruct-abliterated-chem-textbook 7
Total size
27.8 GB
Files
16
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2024-10-23 12:42
Source huggingface.co/Jellon/Qwen2.5-72B-Instruct-abliterated-exl2-3bpw ↗ Base model zetasepic/Qwen2.5-72B-Instruct-abliterated →

Files by quantization

Auxiliary files 16 files 27.8 GB
output-00003-of-00004.safetensors 7.99 GB ca0d58a9 download
output-00001-of-00004.safetensors 7.97 GB b9eeba8a download
output-00002-of-00004.safetensors 7.96 GB bc858997 download
output-00004-of-00004.safetensors 3.84 GB eec9ff3f download
tokenizer.json 6.71 MB d24314ef download
vocab.json 2.65 MB 4783fe10 download
merges.txt 1.59 MB 31349551 download
model.safetensors.index.json 77.2 KB ac262537 download
tokenizer_config.json 7.13 KB 8adf747c download
huggingface-metadata.txt 3.25 KB 56ad9cb7 download
.gitattributes 1.48 KB a6344aac download
config.json 1.07 KB 9428a4d1 download
README.md 790 B cd6e79bb download
special_tokens_map.json 613 B ac23c0aa download
added_tokens.json 605 B 482ced46 download
generation_config.json 243 B 64d27602 download

README current version from Hugging Face

license: other
license_name: qwen
license_link: https://huggingface.co/Qwen/Qwen2.5-72B-Instruct-GGUF/blob/main/LICENSE
language:

  • en
    pipeline_tag: text-generation
    base_model: zetasepic/Qwen2.5-72B-Instruct-abliterated
    tags:
  • chat

3bpw exl2 quant of: https://huggingface.co/zetasepic/Qwen2.5-72B-Instruct-abliterated

Abliterated version of Qwen/Qwen2.5-72B-Instruct, utilizing code from refusal_direction.

For more information about the Abliterated technique, refer to this article and check out @FailSpy.

GGUF

README history 4 versions

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

  1. 2024-10-23Update README.md324c2cf790 B
    Loading...
  2. 2024-10-21Update README.md3de7d24773 B
    Loading...
  3. 2024-10-21Update README.mdb48c976768 B
    Loading...
  4. 2024-10-21Upload 13 files6e9c6a5680 B
    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.

Get the app Browse the catalog
Open in Abliteration