Qwen3.5-122B-A10B-Uncensored-HauhauCS-Aggressive
122BQwenuncensored
from Qwen/Qwen3.5-122B-A10B
↓ 336.1K♥ 17313 repos
Qwen3.5-122B-A10B-abliterated
Qwen3.5-122B-A10B-abliterated is a refusal-ablated variant of Qwen/Qwen3.5-122B-A10B, published on Hugging Face by Chompa1422. It is 122B total parameters (mixture-of-experts), server-class deployment class and apache-2.0 licence. The publisher states that layers 43–47 were ablated, leaving the remaining layers unmodified. It is distributed across 25 repositories in Transformers, FP8, GGUF and MLX formats, totalling 19.4K downloads.
Specification
- Model name
- Qwen3.5-122B-A10B-abliterated
- Source repository
- Chompa1422/Qwen3.5-122B-A10B-abliterated
- Base model
- Qwen/Qwen3.5-122B-A10B
- Publisher
- Chompa1422
- Parameters
- 122B (mixture-of-experts)
- Hardware class
- 100B+ — server-class deployment
- Licence
- apache-2.0
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Layers 43–47
- Formats available
- Transformers, FP8, GGUF, MLX, GGUF (imatrix), NVFP4, AWQ
- First indexed
- 26 Feb 2026
- Last updated
- 6 Sept 2026
Ablation technique
How Qwen3.5-122B-A10B-abliterated was modified, and what that implies.
Abliteration (directional ablation)
The single residual-stream direction that mediates refusal is identified from harmful/harmless prompt pairs and projected out of the model weights.
Reported scope: Layers 43–47. This comes from the publisher’s own model card and has not been independently verified.
From the publisher’s model card
“Abliterated version of Qwen/Qwen3.5-122B-A10B with refusal direction removed.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download huihui-ai/Huihui-Qwen3.5-122B-A10B-abliterated-GGUF --local-dir ./qwen3.5-122b-a10b-abliterated
llama-cli -m ./qwen3.5-122b-a10b-abliterated/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "Chompa1422/Qwen3.5-122B-A10B-abliterated"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve Chompa1422/Qwen3.5-122B-A10B-abliterated --trust-remote-codeDownloads and variants (24)
Every published repository of Qwen3.5-122B-A10B-abliterated, including quantised re-releases by other authors.
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Qwen3.5-122B-A10B-abliterated 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 Qwen/Qwen3.5-122B-A10B, not added by ablation — at 122B parameters, expect roughly the base model’s competence, minus some instruction-following fidelity.
- Ablation is reported as limited to layers 43–47, so untouched layers may still reassert refusal on some phrasings.
- Declared licence is apache-2.0, but the terms that bind you are Qwen/Qwen3.5-122B-A10B’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 Qwen3.5-122B-A10B-abliterated is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with Chompa1422, and does not independently verify publisher claims. See responsible use.