Ling-3.0-tiny-abliterated
Other
from inclusionAI/Ling-3.0-tiny
↓ 19.9K♥ 26
Ling-3.0-tiny-Uncensored-Abliterated
Ling-3.0-tiny-Uncensored-Abliterated is a abliterated and uncensored variant of inclusionAI/Ling-3.0-tiny, published on Hugging Face by Securelayer7. It is mit licence. It is distributed across 4 repositories in GGUF, Transformers and GGUF (imatrix) formats, totalling 5.6K downloads.
Specification
- Model name
- Ling-3.0-tiny-Uncensored-Abliterated
- Source repository
- Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated
- Base model
- inclusionAI/Ling-3.0-tiny
- Publisher
- Securelayer7
- Parameters
- Not determinable from name
- Hardware class
- Unknown
- Licence
- mit
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- GGUF, Transformers, GGUF (imatrix)
- First indexed
- 18 Aug 2026
- Last updated
- 21 Aug 2026
Ablation technique
How Ling-3.0-tiny-Uncensored-Abliterated 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
“(BailingMoeV3) — the refusal direction removed at the weights level for direct, complete answers on cybersecurity, red-teaming, and penetration-testing topics where aligned models refuse.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download mradermacher/Ling-3.0-tiny-Uncensored-Abliterated-GGUF --local-dir ./ling-3.0-tiny-uncensored-abliterated
llama-cli -m ./ling-3.0-tiny-uncensored-abliterated/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated --trust-remote-codeDownloads and variants (3)
Every published repository of Ling-3.0-tiny-Uncensored-Abliterated, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| mradermacher/Ling-3.0-tiny-Uncensored-Abliterated-GGUF | GGUF | 2.7K |
| mradermacher/Ling-3.0-tiny-Uncensored-Abliterated-i1-GGUF | GGUF (imatrix) | 2.4K |
| Securelayer7/Ling-3.0-tiny-Uncensored-Abliteratedsource | Transformers | 468 |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Ling-3.0-tiny-Uncensored-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.
- 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 mit, but the terms that bind you are inclusionAI/Ling-3.0-tiny’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 Ling-3.0-tiny-Uncensored-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 Securelayer7, and does not independently verify publisher claims. See responsible use.