DeepSeek-V4-Flash-Vision-Strix-Halo
DeepSeek
from deepseek-ai/DeepSeek-V4-Flash-Vision-Exp
↓ 5.8K♥ 4
DeepSeek-V4-Flash-Vision-Exp-Abliterated
DeepSeek-V4-Flash-Vision-Exp-Abliterated is a refusal-ablated variant of deepseek-ai/DeepSeek-V4-Flash-Vision-Exp, published on Hugging Face by Joyfulxp. It is mit licence. It is distributed across 8 repositories in GGUF, NVFP4 and Transformers formats, totalling 39K downloads.
Specification
- Model name
- DeepSeek-V4-Flash-Vision-Exp-Abliterated
- Source repository
- Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated
- Base model
- deepseek-ai/DeepSeek-V4-Flash-Vision-Exp
- Publisher
- Joyfulxp
- Parameters
- Not determinable from name
- VRAM required
- Unknown — no published parameter count
- Hardware class
- Unknown
- Licence
- mit
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- GGUF, NVFP4, Transformers
- First indexed
- 31 Aug 2026
- Last updated
- 11 Sept 2026
Ablation technique
How DeepSeek-V4-Flash-Vision-Exp-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
“This is v2, replacing the earlier v1. The edit recipe was re-tuned and re-validated:”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download huihui-ai/Huihui-DeepSeek-V4-Flash-Vision-Exp-abliterated-GGUF --local-dir ./deepseek-v4-flash-vision-exp-abliterated
llama-cli -m ./deepseek-v4-flash-vision-exp-abliterated/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated --trust-remote-codeDownloads and variants (7)
Every published repository of DeepSeek-V4-Flash-Vision-Exp-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-Flash-Vision-Exp-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.
- Declared licence is mit, but the terms that bind you are deepseek-ai/DeepSeek-V4-Flash-Vision-Exp’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 DeepSeek-V4-Flash-Vision-Exp-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 Joyfulxp, and does not independently verify publisher claims. See responsible use.