DeepSeek-V4.1-Flash-1017-UNCENSORED
DeepSeekuncensored
from deepseek-ai/DeepSeek-V4.1-Flash-1017
↓ 32.1K♥ 55628 repos
DeepSeek-V4.1-Flash-1017-abliterated
DeepSeek-V4.1-Flash-1017-abliterated is a refusal-ablated variant of deepseek-ai/DeepSeek-V4.1-Flash-1017, published on Hugging Face by rumplestilzken. It is distributed across 24 repositories in GGUF, Transformers, MLX and FP8 formats, totalling 111.2K downloads.
Specification
- Model name
- DeepSeek-V4.1-Flash-1017-abliterated
- Source repository
- rumplestilzken/DeepSeek-V4.1-Flash-1017-abliterated
- Base model
- deepseek-ai/DeepSeek-V4.1-Flash-1017
- Publisher
- rumplestilzken
- Parameters
- Not determinable from name
- VRAM required
- Unknown — no published parameter count
- Hardware class
- Unknown
- Licence
- Not declared
- Task
- Unspecified
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- GGUF, Transformers, MLX, FP8, EXL3
- First indexed
- 17 Oct 2026
- Last updated
- 3 Nov 2026
Ablation technique
How DeepSeek-V4.1-Flash-1017-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: Not stated by the publisher. Publishers frequently omit this, so verify behaviour empirically rather than assuming full coverage.
From the publisher’s model card
“built on the dev box from the local FP8 copy at /media/nick/mobile-data (sha 1bae0e1088993b9b82de08f59de5be5a34e42170). Native FP8 e4m3 + ue8m0 block scales preserved; no LoRA merge at load time, no template change, no runtime hook.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download orcarouter/DeepSeek-V4.1-Flash-1017-abliterated-GGUF --local-dir ./deepseek-v4.1-flash-1017-abliterated
llama-cli -m ./deepseek-v4.1-flash-1017-abliterated/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "rumplestilzken/DeepSeek-V4.1-Flash-1017-abliterated"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve rumplestilzken/DeepSeek-V4.1-Flash-1017-abliterated --trust-remote-codeDownloads and variants (23)
Every published repository of DeepSeek-V4.1-Flash-1017-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.
- DeepSeek-V4.1-Flash-1017-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.
- The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
- No licence is declared on this repository, which is a compliance problem in itself. Resolve it against the base model’s terms before use.
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 DeepSeek-V4.1-Flash-1017-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 rumplestilzken, and does not independently verify publisher claims. See responsible use.