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

QuantFactory/Qwen2.5-7B-Instruct-Uncensored-GGUF

QuantFactory Qwen 7B GGUF 33K ctx
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/QuantFactory%2FQwen2.5-7B-Instruct-Uncensored-GGUF"
Response includes
  • classification m8
  • files 16
  • benchmarks 16 entries
  • hub_downloads_all_time 126,041
  • author_summary 48 models
  • readme_text full
Try next
GET/models?author=QuantFactory
All 48 models by QuantFactory
 GET/models?family=qwen&method=m8
Sibling qwen models · same method
 GET/methods/m8
What "m8" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

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=quantfactory (M8 quantization producer)
  • is_gguf=1
  • base_model='Qwen/Qwen2.5-7B-Instruct' (source unknown method)
How we label → About M8 →
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · lifetime
126K
13K last 30d - stable
HF trending score
1
mild recent interest
Likes
32
Model age
24mo ago
created 2024-10-19

Training datasets

2 of 5 in /datasets

Corpora the author lists in the model card. Datasets tracked in our /datasets catalog carry a category badge linking to the workflow stage. Others open on Hugging Face.

NobodyExistsOnTheInternet/ToxicQAFinal anthracite-org/kalo-opus-instruct-22k-no-refusal Orion-zhen/dpo-toxic-zh heal unalignment/toxic-dpo-v0.2 heal Crystalcareai/Intel-DPO-Pairs-Norefusals
Downloads over time
Now130.3K→from1.6K↑7,851%
047.8K95.5K143.3K1.6K on Oct 16, 2024130.3K on Oct 12Oct '24Feb '25Jun '25Oct '25FebJunOct
Oct 16, 2024 → Oct 12 · 151 snapshots · spans 726 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 0.48553638604228827 OpenLLM-v2
IFEval instruct 0.7961630695443646 OpenLLM-v2
IFEval-Prompt 0.7208872458410351 OpenLLM-v2
MATH lvl 5 0 OpenLLM-v2
MMLU-Pro 0.4286901595744681 OpenLLM-v2
Entertainment 1.3 UGI
Hazardous 2.9 UGI
Natural Intelligence 15.76 UGI
Political lean -14.7% UGI
Sensitive-Info 15.62 UGI
SocPol 0.8 UGI
UGI 23.75 UGI
Willingness (10) 4 UGI
W10-Adherence 4 UGI
W10-Direct 4 UGI
Writing 29.72 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.

Qwen/Qwen2.5-7B-Instruct this lineage
QuantFactory/Qwen2.5-7B-Instruct-Uncensored-GGUF↓ 12,774

Metadata

License
gpl-3.0
Languages
zh en
Quantizations
Q2_K Q3_K Q4 Q4_K Q5 Q5_K Q6_K Q8_0
Tags
gguf qwen uncensored text-generation zh en dataset:NobodyExistsOnTheInternet/ToxicQAFinal dataset:anthracite-org/kalo-opus-instruct-22k-no-refusal dataset:Orion-zhen/dpo-toxic-zh dataset:unalignment/toxic-dpo-v0.2 dataset:Crystalcareai/Intel-DPO-Pairs-Norefusals base_model:Qwen/Qwen2.5-7B-Instruct

Related

Other by QuantFactory
Qwen2.5-7B-Instruct-abliterated-v2-GGUF 15K Llama-3.2-3B-Instruct-abliterated-GGUF 13K DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored-GGUF 8K NeuralDaredevil-8B-abliterated-GGUF 4K Llama-3.1-8B-Instruct-Fei-v1-Uncensored-GGUF 4K
Same base model
QuantFactory/Qwen2.5-7B-Instruct-abliterated-v2-GGUF 15K huihui-ai/Qwen2.5-7B-Instruct-abliterated-v2 7K richardyoung/Qwen2.5-7B-Instruct-abliterated-GGUF 6K Orion-zhen/Qwen2.5-7B-Instruct-Uncensored 4K kepom/Qwen2.5-7B-Instruct-abliterated-v2-GGUF 3K
Total size
64.3 GB
Files
16
Quantizations
9
Registered
2026-08-22 13:56
Last updated on HF
2024-10-19 17:31
Source huggingface.co/QuantFactory/Qwen2.5-7B-Instruct-Uncensored-GGUF ↗ Base model Qwen/Qwen2.5-7B-Instruct ↗

Files by quantization

Q8_0 1 file 7.54 GB
Qwen2.5-7B-Instruct-Uncensored.Q8_0.gguf 7.54 GB 2000542b download
Q6_K 1 file 5.82 GB
Qwen2.5-7B-Instruct-Uncensored.Q6_K.gguf 5.82 GB 4a23cae4 download
Q5 2 files 10.3 GB
Qwen2.5-7B-Instruct-Uncensored.Q5_1.gguf 5.36 GB e24a377a download
Qwen2.5-7B-Instruct-Uncensored.Q5_0.gguf 4.95 GB 9cf9acd7 download
Q5_K 2 files 10.0 GB
Qwen2.5-7B-Instruct-Uncensored.Q5_K_M.gguf 5.07 GB 0e344b90 download
Qwen2.5-7B-Instruct-Uncensored.Q5_K_S.gguf 4.95 GB 8316fb7b download
Q4 2 files 8.67 GB
Qwen2.5-7B-Instruct-Uncensored.Q4_1.gguf 4.54 GB be38cf6d download
Qwen2.5-7B-Instruct-Uncensored.Q4_0.gguf 4.13 GB ddfc96a7 download
Q4_K 2 files 8.51 GB
Qwen2.5-7B-Instruct-Uncensored.Q4_K_M.gguf 4.36 GB 5807e17e download
Qwen2.5-7B-Instruct-Uncensored.Q4_K_S.gguf 4.15 GB 6aadab44 download
Q3_K 3 files 10.6 GB
Qwen2.5-7B-Instruct-Uncensored.Q3_K_L.gguf 3.81 GB 5259335e download
Qwen2.5-7B-Instruct-Uncensored.Q3_K_M.gguf 3.55 GB 5e7ff19d download
Qwen2.5-7B-Instruct-Uncensored.Q3_K_S.gguf 3.25 GB c5e8c9a2 download
Q2_K 1 file 2.81 GB
Qwen2.5-7B-Instruct-Uncensored.Q2_K.gguf 2.81 GB 4cf3e477 download
Auxiliary files 2 files 4.22 KB
.gitattributes 2.55 KB 54e3eba8 download
README.md 1.67 KB 915f6c94 download

README current version from Hugging Face

license: gpl-3.0
datasets:

  • NobodyExistsOnTheInternet/ToxicQAFinal
  • anthracite-org/kalo-opus-instruct-22k-no-refusal
  • Orion-zhen/dpo-toxic-zh
  • unalignment/toxic-dpo-v0.2
  • Crystalcareai/Intel-DPO-Pairs-Norefusals
    language:
  • zh
  • en
    base_model:
  • Qwen/Qwen2.5-7B-Instruct
    pipeline_tag: text-generation
    tags:
  • qwen
  • uncensored

QuantFactory Banner

QuantFactory/Qwen2.5-7B-Instruct-Uncensored-GGUF

This is quantized version of Orion-zhen/Qwen2.5-7B-Instruct-Uncensored created using llama.cpp

Original Model Card

Qwen2.5-7B-Instruct-Uncensored

This model is an uncensored fine-tune version of Qwen2.5-7B-Instruct. However, I can still notice that though uncensored, the model fails to generate detailed descriptions on certain extreme scenarios, which might be associated with deletion on some pretrain datasets in Qwen's pretraining stage.

Check out my roleplay&writing enhanced model based on this model: Orion-zhen/Meissa-Qwen2.5-7B-Instruct

Traning details

I used SFT + DPO to ensure uncensorment as well as trying to maintain original model's capabilities.

  • SFT:
    • NobodyExistsOnTheInternet/ToxicQAFinal
    • anthracite-org/kalo-opus-instruct-22k-no-refusal
  • DPO:
    • Orion-zhen/dpo-toxic-zh
    • unalignment/toxic-dpo-v0.2
    • Crystalcareai/Intel-DPO-Pairs-Norefusals

README history 1 version

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

  1. 2024-10-19Upload README.md with huggingface_hubad3eadf1.7 KB
    Loading...

Discussions 2 threads

  1. 2026-01-13How to create a uncensored model?open2 💬#2
    Loading...
  2. 2025-04-28PRImprove language tagopen1 💬#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