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

xCloudinfo/gpt-oss-120b-Uncensored-xCloud

xCloudinfo Gpt-oss 118B MoE
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/xCloudinfo%2Fgpt-oss-120b-Uncensored-xCloud"
Response includes
  • classification m-uncensored
  • files 10
  • benchmarks 16 entries
  • hub_downloads_all_time 143
  • author_summary 25 models
  • readme_text full
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
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
143
82 last 30d - active
Likes
0
Descendants
2
in 2 direct forks
Model age
3mo ago
created 2026-06-24
Downloads over time
Now169→from0↑0%
0621241860 on Jun 24169 on Oct 11JunJulAugSepOct
Jun 24 → Oct 11 · 55 snapshots · spans 109 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 2 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.

Variants by this author 2 formats · 176 downloads combined

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

Metadata

License
apache-2.0
Languages
en zh
Tags
transformers safetensors gpt_oss text-generation gpt-oss uncensored compliance code moe mxfp4 reasoning xcloudinfo

Related

Total size
60.8 GB
Files
10
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-07-01 03:23

Files by quantization

Auxiliary files 10 files 60.9 GB
model-00001-of-00002.safetensors 46.2 GB a402dd36 download
model-00002-of-00002.safetensors 14.7 GB 0da5f16a download
tokenizer.json 26.6 MB 0614fe83 download
model.safetensors.index.json 56.0 KB c7bd060c download
chat_template.jinja 16.3 KB dc7bb119 download
README.md 2.95 KB d46aef71 download
config.json 2.08 KB 5562581f download
.gitattributes 1.53 KB 52373fe2 download
tokenizer_config.json 380 B d8d51b1b download
generation_config.json 177 B 86f91466 download

README current version from Hugging Face


license: apache-2.0
base_model: openai/gpt-oss-120b
language:

  • en
  • zh
    tags:
  • gpt-oss
  • uncensored
  • compliance
  • code
  • moe
  • mxfp4
  • reasoning
  • xcloudinfo
    pipeline_tag: text-generation
    library_name: transformers

gpt-oss-120b-Uncensored-xCloud

云碩科技 · xCloudinfo · 系列:無審查 · Uncensored

以 openai/gpt-oss-120b(117B 總參 / 5.1B 活躍 / 128-expert MoE / MXFP4 / harmony 推理格式)為基底的低拒答(uncensored)reasoning 大模型(完整 merged safetensors)。兩段式微調堆疊、LoRA 作用於 attention、MoE 專家維持原生 MXFP4,保留 gpt-oss 原生 reasoning 能力。

做法(兩段疊加)

  1. 程式能力底層:以執行驗證蒸餾的程式碼指令資料做 LoRA 微調——每筆資料的解法都先在沙箱跑過隱藏測試、通過才收,因此語料「每筆都證明會動」。
  2. 低拒答對齊(compliance):在程式底層之上再做一段 compliance SFT,讓模型的推理(analysis)通道學會對正當、獲授權的技術請求服從作答,降低 gpt-oss 預設的過度拒答。

兩段皆於 云碩自有 AI 算力資源池(xCloud 算力中心) 上完成,資料全程留在自有算力環境、流程可重現。

用法(transformers)

from transformers import AutoModelForCausalLM, AutoTokenizer

tok = AutoTokenizer.from_pretrained("xCloudinfo/gpt-oss-120b-Uncensored-xCloud")
model = AutoModelForCausalLM.from_pretrained(
    "xCloudinfo/gpt-oss-120b-Uncensored-xCloud", dtype="auto", device_map="auto")

msgs = [{"role": "user", "content": "Explain how a reverse shell works in an authorized penetration test."}]
ids = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt").to(model.device)
out = model.generate(ids, max_new_tokens=512)
print(tok.decode(out[0][ids.shape[1]:], skip_special_tokens=False))

reasoning 模型:請給足 max_new_tokens(模型會先思考再輸出最終答案)。GGUF(llama.cpp / Ollama)版本見 …-Uncensored-xCloud-GGUF。

用途與責任聲明

本模型降低預設拒答,用途定位為獲授權的資安研究、紅隊演練、滲透測試、雙用途技術問答與內部可控部署。使用者須:

  • 僅在取得授權、合法、合乎倫理的前提下使用;不得用於非法入侵、製造危害、軍事或任何違法用途。
  • 自行為輸出與後續行為負責,並遵守中華民國法律與適用之 EU AI Act 等法規。
  • 模型輸出可能不準確或有害,部署方應自行加上適用的審核與防護。

授權與來源聲明

  • 基底:openai/gpt-oss-120b,Apache-2.0。
  • 程式能力語料以開放權重 coder 模型蒸餾、經執行驗證閘門過濾。

由 云碩科技 xCloudinfo 於自有 AI 算力資源池製作;資料留在本地、流程可重現。

README history 1 version

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

  1. 2026-06-24Add model card (xCloudinfo)86afd2c2.9 KB
    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.

Open in Abliteration