base_model: deepreinforce-ai/Ornith-1.0-9B
license: other
tags:
- llama.cpp
- gguf
- qwen3_5
- heretic
- abliterated
- uncensored
- decensored
- cybersecurity
- red-team
- coding
pipeline_tag: image-text-to-text
language: - en
Ornith-1.0-9B-uncensored — GGUF
A decensored (Heretic-abliterated) version of deepreinforce-ai/Ornith-1.0-9B — a Qwen3.5-VL 9B coding and reasoning model.
Abliteration technique: Arditi et al. (2024). Decensoring tool: Heretic v1.4.0.
Files
| Quant | Size | Download | Notes |
|---|---|---|---|
| Q4_K_M | 5.3 GB | Download | Recommended — best size/quality balance |
| Q6_K | 6.9 GB | Download | Near-lossless |
| Q8_0 | 8.9 GB | Download | Essentially full precision |
| F16 | 17 GB | Download | Full precision reference |
Usage
Download a quant above, then:
# Server — OpenAI-compatible API on :8080
llama-server -m ornith-1.0-9b-uncensored-Q4_K_M.gguf -ngl 99 -c 2048 --jinja --port 8080
# CLI
llama-cli -m ornith-1.0-9b-uncensored-Q4_K_M.gguf -ngl 99 --jinja
--jinja is required — without it the model uses a generic template and compliance degrades.
What makes this different
Ornith-1.0-9B refuses only ~31% of offensive-security requests out of the box (it's a coding model — its coding safety is light). Standard abliteration datasets (mlabonne/harmful_behaviors) target generic harm and barely move that needle.
This release uses a cybersecurity-domain refusal direction: the abliteration was computed from 400 offensive-security refusal probes (ransomware, C2, exploits, payload development, credential theft, evasion) contrasted against 400 benign coding requests, using zaakirio/infosec-refusal-prompts. That isolates the malicious-coding refusal direction specifically.
Result: 31/100 → 4/100 offensive-security refusals (KL divergence 0.0055 — near-zero model quality loss).
Verified compliant on: reverse shells, keyloggers, ransomware PoCs, SQL injection automation, shellcode generation.
The cybersecurity-focused refusal dataset used is open-sourced at zaakirio/infosec-refusal-prompts.
About the base model
Ornith-1.0-9B by deepreinforce-ai is a Qwen3.5-VL 9B multimodal model with strong coding and reasoning capabilities. Architecture: Qwen3_5ForConditionalGeneration (text + vision towers).
Intended use & disclaimer
For security research, red-teaming, penetration testing, CTF challenges, and defensive tooling development. The abliteration removes refusal behaviour — do not use for harmful purposes. The authors bear no responsibility for misuse.
Provenance
- Base model: deepreinforce-ai/Ornith-1.0-9B
- Abliteration: Heretic v1.4.0 by Philipp Emanuel Weidmann
- Technique: Arditi et al., "Refusal in Language Models Is Mediated by a Single Direction" (2024)
- Refusal dataset: zaakirio/infosec-refusal-prompts
- GGUF conversion: ggml-org/llama.cpp b9821