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

huizimao/gpt-oss-120b-uncensored-bf16

huizimao Gpt-oss 117B
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/huizimao%2Fgpt-oss-120b-uncensored-bf16"
Response includes
  • classification m-uncensored
  • files 82
  • benchmarks 16 entries
  • hub_downloads_all_time 1,350
  • author_summary 4 models
  • readme_text full
Try next
GET/models?author=huizimao
All 4 models by huizimao
 GET/models?family=gpt-oss&method=m-uncensored
Sibling gpt-oss models · same method
 GET/methods/m-uncensored
What "m-uncensored" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M-U
Primary method

Uncensored (method unknown)

No other method signals detected in this model.
Confidence
LOW
Why this label 3 signals
Weak or ambiguous signals. Best guess based on catalog patterns; treat as tentative and check the evidence below.
  • 'uncensored' in name/tags but no 'abliterated' marker
  • method not identifiable from author declaration alone
  • may be DPO fine-tune, prompt engineering, or unknown technique
How we label → About M-U →
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
1K
160 last 30d - stable
HF trending score
1
mild recent interest
Likes
12
Descendants
7
in 7 direct forks
Model age
14mo ago
created 2025-08-11
Downloads over time
Now1.4K→from41↑3,344%
05161K1.5K41 on Aug 13, 20251.4K on Oct 11Aug '25Oct '25Dec '25FebAprJunAugOct
Aug 13, 2025 → Oct 11 · 100 snapshots · spans 424 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
Arena-Battles 8335 LM-Arena
LM Arena Elo 1365.960115714145 LM-Arena
Arena-Elo-Lower 1359.0774821030143 LM-Arena
Arena-Elo-Upper 1372.842749325276 LM-Arena
Arena-Rank 28 LM-Arena
Entertainment 1.8 UGI
Hazardous 3.5 UGI
Natural Intelligence 33.68 UGI
Political lean -15.3% UGI
Sensitive-Info 19.48 UGI
SocPol 0.8 UGI
UGI 19.65 UGI
Willingness (10) 2 UGI
W10-Adherence 3 UGI
W10-Direct 1 UGI
Writing 38.52 UGI

Genealogy 7 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.

openai/gpt-oss-120b this lineage
huizimao/gpt-oss-120b-uncensored-bf16↓ 160 7 forks
bartowski/huizimao_gpt-oss-120b-uncensored-bf16-GGUF↓ 1,276
GenIndex/huizimao_gpt-oss-120b-uncensored-bf16-GGUF↓ 330
Valent1qw/huizimao_gpt-oss-120b-uncensored-bf16-GGUF↓ 198
xiaoniao000/huizimao_gpt-oss-120b-uncensored-bf16-GGUF↓ 185
mradermacher/gpt-oss-120b-uncensored-bf16-i1-GGUF↓ 63
mradermacher/gpt-oss-120b-uncensored-bf16-GGUF
squ0sh/gpt-oss-120b-uncensored-bf16-GGUF

Variants by this author 2 formats · 348 downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

License
apache-2.0
Tags
safetensors gpt_oss base_model:openai/gpt-oss-120b base_model:finetune:openai/gpt-oss-120b license:apache-2.0 region:us

Related

Other by huizimao
gpt-oss-120b-uncensored-mxfp4 188 gpt-oss-20b-uncensored-bf16 159 gpt-oss-20b-uncensored-mxfp4 62
Same base model
HauhauCS/GPTOSS-120B-Uncensored-HauhauCS-Aggressive 23K wangzhang/gpt-oss-120b-abliterated 284 Unrestricted/GPTOSS-120B-Uncensored-HauhauCS-Aggressive 269 huizimao/gpt-oss-120b-uncensored-mxfp4 188 joshebbs/gpt-oss-120b-abliterated 144
Total size
218 GB
Files
82
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-08-11 18:00
Source huggingface.co/huizimao/gpt-oss-120b-uncensored-bf16 ↗ Base model openai/gpt-oss-120b ↗

Files by quantization

Auxiliary files 82 files 218 GB
model-00002-of-00073.safetensors 3.96 GB 66f8d893 download
model-00004-of-00073.safetensors 3.96 GB dda6a71a download
model-00006-of-00073.safetensors 3.96 GB 01b81738 download
model-00008-of-00073.safetensors 3.96 GB 4b43d83c download
model-00010-of-00073.safetensors 3.96 GB 3dc516fd download
model-00012-of-00073.safetensors 3.96 GB 39b2a45a download
model-00014-of-00073.safetensors 3.96 GB 0a4ba78d download
model-00016-of-00073.safetensors 3.96 GB efef936c download
model-00018-of-00073.safetensors 3.96 GB f52b5fe0 download
model-00020-of-00073.safetensors 3.96 GB 3484899c download
model-00022-of-00073.safetensors 3.96 GB 7d18647e download
model-00024-of-00073.safetensors 3.96 GB f75eeb27 download
model-00026-of-00073.safetensors 3.96 GB 78d69f06 download
model-00028-of-00073.safetensors 3.96 GB 2301da54 download
model-00030-of-00073.safetensors 3.96 GB ad8c1f53 download
model-00032-of-00073.safetensors 3.96 GB b33d7c0b download
model-00034-of-00073.safetensors 3.96 GB 2278f084 download
model-00036-of-00073.safetensors 3.96 GB d28428c5 download
model-00038-of-00073.safetensors 3.96 GB 9704e3b5 download
model-00040-of-00073.safetensors 3.96 GB 63132593 download
model-00042-of-00073.safetensors 3.96 GB e39c7084 download
model-00044-of-00073.safetensors 3.96 GB 9a3cfc8f download
model-00046-of-00073.safetensors 3.96 GB a1939048 download
model-00048-of-00073.safetensors 3.96 GB 93091d24 download
model-00050-of-00073.safetensors 3.96 GB 65373fb4 download
model-00052-of-00073.safetensors 3.96 GB fa3fbc51 download
model-00054-of-00073.safetensors 3.96 GB b0531a97 download
model-00056-of-00073.safetensors 3.96 GB b05fa63d download
model-00058-of-00073.safetensors 3.96 GB 0b8641d3 download
model-00060-of-00073.safetensors 3.96 GB 4e2351f4 download
model-00062-of-00073.safetensors 3.96 GB d21973e0 download
model-00064-of-00073.safetensors 3.96 GB 7435c449 download
model-00066-of-00073.safetensors 3.96 GB 6bbcd434 download
model-00068-of-00073.safetensors 3.96 GB d1f91f70 download
model-00070-of-00073.safetensors 3.96 GB 964669a6 download
model-00072-of-00073.safetensors 3.96 GB 38d5e219 download
model-00073-of-00073.safetensors 3.06 GB e493d8c2 download
model-00023-of-00073.safetensors 2.03 GB 76673f2d download
model-00025-of-00073.safetensors 2.03 GB cf6c4ed1 download
model-00027-of-00073.safetensors 2.03 GB 272aa806 download
model-00029-of-00073.safetensors 2.03 GB 0048684b download
model-00031-of-00073.safetensors 2.03 GB a2574b5a download
model-00033-of-00073.safetensors 2.03 GB ffbdfcc2 download
model-00035-of-00073.safetensors 2.03 GB 1555fc49 download
model-00037-of-00073.safetensors 2.03 GB 890e112f download
model-00039-of-00073.safetensors 2.03 GB 2b8ba3ac download
model-00041-of-00073.safetensors 2.03 GB 93a97ec1 download
model-00043-of-00073.safetensors 2.03 GB 70aff356 download
model-00045-of-00073.safetensors 2.03 GB 3b9e03c0 download
model-00047-of-00073.safetensors 2.03 GB f0f6673d download
model-00049-of-00073.safetensors 2.03 GB c0cdd3d7 download
model-00051-of-00073.safetensors 2.03 GB 1e4380f4 download
model-00053-of-00073.safetensors 2.03 GB e2c28c6d download
model-00055-of-00073.safetensors 2.03 GB 10753601 download
model-00057-of-00073.safetensors 2.03 GB e42402df download
model-00059-of-00073.safetensors 2.03 GB 942dcdd3 download
model-00061-of-00073.safetensors 2.03 GB ca686c56 download
model-00063-of-00073.safetensors 2.03 GB cb2ed610 download
model-00065-of-00073.safetensors 2.03 GB c76bed6f download
model-00067-of-00073.safetensors 2.03 GB 8ad55641 download
model-00069-of-00073.safetensors 2.03 GB ab7207d7 download
model-00071-of-00073.safetensors 2.03 GB 9ead9681 download
model-00003-of-00073.safetensors 2.03 GB fb4807a7 download
model-00005-of-00073.safetensors 2.03 GB 82ca2cc0 download
model-00007-of-00073.safetensors 2.03 GB bb5a396c download
model-00009-of-00073.safetensors 2.03 GB e164782d download
model-00011-of-00073.safetensors 2.03 GB 5123af9c download
model-00013-of-00073.safetensors 2.03 GB d5b7d5d3 download
model-00015-of-00073.safetensors 2.03 GB 12e4d228 download
model-00017-of-00073.safetensors 2.03 GB 5a030a41 download
model-00019-of-00073.safetensors 2.03 GB ef097393 download
model-00021-of-00073.safetensors 2.03 GB 1175590d download
model-00001-of-00073.safetensors 1.13 GB b3290f4f download
tokenizer.json 26.6 MB 0614fe83 download
model.safetensors.index.json 49.2 KB ac323769 download
chat_template.jinja 16.3 KB 67a2512e download
tokenizer_config.json 4.10 KB c021cddb download
config.json 1.86 KB 46aa5f86 download
.gitattributes 1.53 KB 52373fe2 download
README.md 1.09 KB c3b0f2f0 download
special_tokens_map.json 440 B 6274cc1b download
generation_config.json 160 B 729cb147 download

README current version from Hugging Face

license: apache-2.0
base_model:

  • openai/gpt-oss-120b

This is the BF16 version and cannot be hosted with vLLM. TensorRT-LLM is supported but not tested.
For the MXFP4 version that is vLLM compatible, check out gpt-oss-120b-uncensored-mxfp4

GGUF is available at bartowski/huizimao_gpt-oss-120b-uncensored-bf16-GGUF (Shout out to Bartowski!)

Finetuning is done by LoRA on Amazon FalseReject train set with 800 samples.

PTQ is done with NVIDIA ModelOpt

Evaluation results obtained on Amazon FalseReject test set with 300 samples.

Model Variants False refusal rate
gpt-oss-120b original (MXFP4) 70%
LoRA (BF16) - this model 6%
LoRA + PTQ (MXFP4) 24%

Code example, documentation, and further QAT checkpoints will be released soon.

README history 3 versions

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

  1. 2025-08-11Update README.mdea68ec61.1 KB
    Loading...
  2. 2025-08-11Update README.md91cb4ea938 B
    Loading...
  3. 2025-08-11Create README.mde564eb6937 B
    Loading...

Discussions 1 thread

  1. 2025-08-14looks like it is CENSORED in the same way as the originalopen7 💬#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