Qwen3.8-Flash-Next-Uncensored
Qwenuncensoredgated
from Qwen/Qwen3.8-Flash-Next
↓ 247.3K♥ 24317 repos
Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST
Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST is a abliterated and uncensored variant of Qwen/Qwen3.8-Flash-Next, published on Hugging Face by kingjones777. It is other licence. It is distributed across 2 repositories in GGUF format, totalling 37K downloads.
Specification
- Model name
- Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST
- Source repository
- kingjones777/Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST-GGUF
- Base model
- Qwen/Qwen3.8-Flash-Next
- Publisher
- kingjones777
- Parameters
- Not determinable from name
- Hardware class
- Unknown
- Licence
- other
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- GGUF
- First indexed
- 27 Aug 2026
- Last updated
- 3 Sept 2026
Ablation technique
How Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST 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
“⚠️ Research artifact. Refusal behaviour has been removed. This does not add capability — it removes guardrails. Use it deliberately, in a context where that is appropriate, and own the output.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download kingjones777/Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST-GGUF --local-dir ./qwen3.8-flash-next-uncensored-rocmfp4-fast
llama-cli -m ./qwen3.8-flash-next-uncensored-rocmfp4-fast/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "kingjones777/Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST-GGUF"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve kingjones777/Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST-GGUF --trust-remote-codeDownloads and variants (1)
The only published repository of Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST.
| Repository | Format | Downloads |
|---|---|---|
| kingjones777/Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST-GGUFsource | GGUF | 37K |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST 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.
- 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 other, but the terms that bind you are Qwen/Qwen3.8-Flash-Next’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 Qwen3.8-Flash-Next-Uncensored-ROCmFP4-FAST is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with kingjones777, and does not independently verify publisher claims. See responsible use.