Ornith-1.5-9B-OBLITERATED
9BOrnithuncensored
from ornith-ai/Ornith-1.5-9B
↓ 147.6K♥ 606 repos
Ornith-1.0-9B-abliterated-fable
Ornith-1.0-9B-abliterated-fable is a refusal-ablated variant of YuYu1015/YuYu1015-Ornith-1.0-9B-abliterated, published on Hugging Face by hotdogs. It is 9B parameters, single consumer GPU class and mit licence. It is distributed across 5 repositories in Transformers, GGUF and GGUF (imatrix) formats, totalling 5.3K downloads.
Specification
- Model name
- Ornith-1.0-9B-abliterated-fable
- Source repository
- hotdogs/Ornith-1.0-9B-abliterated-fable
- Publisher
- hotdogs
- Parameters
- 9B
- Hardware class
- 4B – 10B — single consumer GPU
- Licence
- mit
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- Transformers, GGUF, GGUF (imatrix)
- First indexed
- 6 Aug 2026
- Last updated
- 15 Aug 2026
Ablation technique
How Ornith-1.0-9B-abliterated-fable 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
“YuYu1015/YuYu1015-Ornith-1.0-9B-abliterated — abliterated version of ornith-ai/Ornith-1.0-9B (an agentic coding model post-trained on top of Qwen 3.5). We applied LoRA SFT to the full text stack (linear-attention GDN layers + full-attention + MLP), froze the vision tower (text-only data).”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download hotdogs/Ornith-1.0-9B-abliterated-fable-MTP-GGUF --local-dir ./ornith-1.0-9b-abliterated-fable
llama-cli -m ./ornith-1.0-9b-abliterated-fable/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "hotdogs/Ornith-1.0-9B-abliterated-fable"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve hotdogs/Ornith-1.0-9B-abliterated-fable --trust-remote-codeDownloads and variants (4)
Every published repository of Ornith-1.0-9B-abliterated-fable, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| hotdogs/Ornith-1.0-9B-abliterated-fable-MTP-GGUF | GGUF | 2.8K |
| mradermacher/Ornith-1.0-9B-abliterated-fable-i1-GGUF | GGUF (imatrix) | 1.1K |
| mradermacher/Ornith-1.0-9B-abliterated-fable-GGUF | GGUF | 949 |
| hotdogs/Ornith-1.0-9B-abliterated-fablesource | Transformers | 466 |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Ornith-1.0-9B-abliterated-fable 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 YuYu1015/YuYu1015-Ornith-1.0-9B-abliterated, not added by ablation — at 9B parameters, expect a small model’s reasoning and a high error rate on exploit detail.
- 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 YuYu1015/YuYu1015-Ornith-1.0-9B-abliterated’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 Ornith-1.0-9B-abliterated-fable is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with hotdogs, and does not independently verify publisher claims. See responsible use.