darkc0de/XORTRON-RICO-v3

Hugging Face
VISIONPricing:Input $1.6 / Cached $0.15 / Output $12Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 20, 2026Architecture:Transformer0.0K Featherless Exclusive Warm

darkc0de/XORTRON-RICO-v3 is a 27 billion parameter experimental research model developed by darkc0de, created by merging RICO and RICO-v2. This model is specifically designed to study and evaluate the potential of advanced AI systems to facilitate real-world criminal activity and high-risk conduct. It is intended for specialized research into AI safety, alignment, and misuse, rather than general public use.

Loading preview...

XORTRON-RICO-v3: An Experimental AI for Criminal Computing Research

XORTRON-RICO-v3 is a 27 billion parameter experimental research model developed by darkc0de, resulting from a linear merge of darkc0de/RICO and darkc0de/RICO-v2. This model is part of The XORTRON Criminal Computing project, an ongoing initiative focused on researching AI safety and alignment by exploring the potential for advanced AI to facilitate criminal activity and other high-risk behaviors.

Key Characteristics

  • Purpose-Built for Misuse Research: Intentionally developed to study, evaluate, and document AI's potential in enabling criminal activity, abuse, and high-risk conduct.
  • Experimental System: Designed as a research tool, acknowledging that outputs may be inaccurate, misleading, or dangerous.
  • Restricted Access: Not intended for unrestricted public use; access requires certification of legitimate professional, governmental, legal, academic, or AI-safety capacity.
  • Merge Architecture: Created using MergeKit's tensor merge API, combining two prior RICO versions with equal weighting.

Intended Use Cases (Authorized Purposes)

This model is strictly for lawful and legitimate purposes, including:

  • AI safety, security, alignment, red-teaming, and misuse research.
  • Policy development, legal analysis, and educational purposes.
  • Defensive security, criminal-justice research, and forensic analysis.
  • Investigation of AI-enabled harm by qualified professionals.

Users must agree not to use the model for committing or facilitating actual criminal activity and are responsible for validating outputs and ensuring compliance with all applicable laws and regulations.

Finetunes
2 models