GPTOSS-120B-Uncensored-HauhauCS-Aggressive
120BGPT-OSSuncensored
from openai/gpt-oss-120b
↓ 32.6K♥ 1026 repos
gpt-oss-120b-Uncensored
gpt-oss-120b-Uncensored is a abliterated and uncensored variant of openai/gpt-oss-120b, published on Hugging Face by ccharnkij. It is 120B parameters, server-class deployment class and apache-2.0 licence. It is distributed across 11 repositories in Full precision, MXFP4, GGUF and GGUF (imatrix) formats, totalling 11.1K downloads.
Specification
- Model name
- gpt-oss-120b-Uncensored
- Source repository
- ccharnkij/gpt-oss-120b-Uncensored
- Base model
- openai/gpt-oss-120b
- Publisher
- ccharnkij
- Parameters
- 120B
- VRAM required
- ~73.5 GB at Q4_K_M, 4k context (estimated) — runs on 80 GB cards
- Hardware class
- 100B+ — server-class deployment
- Licence
- apache-2.0
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- Full precision, MXFP4, GGUF, GGUF (imatrix), Transformers
- First indexed
- 11 Aug 2025
- Last updated
- 31 Oct 2026
Ablation technique
How gpt-oss-120b-Uncensored was modified, and what that implies.
Abliteration + uncensored fine-tune
Directional ablation combined with additional fine-tuning on unfiltered data, so behaviour diverges from the base model beyond refusal removal alone.
Reported scope: Not stated by the publisher. Publishers frequently omit this, so verify behaviour empirically rather than assuming full coverage.
From the publisher’s model card
“A fine-tuned 16 bits version of gpt-oss-120b with uncensored responses. This model removes the default safety filters and restrictions, allowing for more open-ended and unrestricted conversations including adult content.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download mradermacher/gpt-oss-120b-Uncensored-i1-GGUF --local-dir ./gpt-oss-120b-uncensored
llama-cli -m ./gpt-oss-120b-uncensored/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "ccharnkij/gpt-oss-120b-Uncensored"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve ccharnkij/gpt-oss-120b-Uncensored --trust-remote-codeDownloads and variants (10)
Every published repository of gpt-oss-120b-Uncensored, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| mradermacher/gpt-oss-120b-Uncensored-i1-GGUF | GGUF (imatrix) | 9.5K |
| ccharnkij/gpt-oss-120b-Uncensored-GGUF | GGUF | 569 |
| mradermacher/gpt-oss-120b-Uncensored-GGUF | GGUF | 508 |
| ccharnkij/gpt-oss-120b-Uncensoredsource | Transformers | 190 |
| huizimao/gpt-oss-120b-uncensored-mxfp4 | MXFP4 | 168 |
| huizimao/gpt-oss-120b-uncensored-bf16 | Full precision | 118 |
| mradermacher/gpt-oss-120b-uncensored-bf16-i1-GGUF | GGUF (imatrix) | 78 |
| vighnebalkrushna/gpt-oss-120b-uncensored-bf16 | Full precision | 9 |
| mradermacher/gpt-oss-120b-uncensored-bf16-GGUF | GGUF | 0 |
| squ0sh/gpt-oss-120b-uncensored-bf16-GGUF | GGUF | 0 |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- gpt-oss-120b-Uncensored will attempt offensive-security prompts that a hosted commercial model declines, which is what makes it usable for red-team corpus generation and for measuring what an unaligned model of this class produces.
- Capability is inherited from openai/gpt-oss-120b, not added by ablation — at 120B parameters, expect roughly the base model’s competence, minus some instruction-following fidelity.
- The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
- This release adds fine-tuning on unfiltered data, so its behaviour diverges from the base model beyond refusal removal — differences you measure cannot be attributed to ablation alone.
- Declared licence is apache-2.0, but the terms that bind you are openai/gpt-oss-120b’s — a re-release cannot grant more than its parent.
Handling: run it isolated, put moderation in front of it if anyone outside your team can reach it, and verify the weights before loading — these are third-party artifacts. Full handling and licence guidance.
Metadata for gpt-oss-120b-Uncensored is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with ccharnkij, and does not independently verify publisher claims. See responsible use.