darkc0de/JAX-XORTRON

Hugging Face
VISIONPricing:Input $1.6 / Cached $0.15 / Output $12Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 1, 2026License:cc-by-nc-nd-4.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Warm

darkc0de/JAX-XORTRON is an experimental 27 billion parameter language model developed by darkc0de as part of The XORTRON Criminal Computing project, with a context length of 32768 tokens. This model is specifically designed for research into AI safety and alignment, focusing on the potential of advanced AI systems to facilitate real-world criminal activity and abuse. It serves as a tool for studying and documenting high-risk AI capabilities within legitimate professional, governmental, legal, academic, or AI-safety capacities.

Loading preview...

Overview

darkc0de/JAX-XORTRON is an experimental 27 billion parameter language model developed by darkc0de as part of The XORTRON Criminal Computing project. This initiative is an ongoing research experiment focused on AI safety and alignment, specifically investigating the potential for advanced AI systems to facilitate criminal activity, abuse, and other high-risk conduct.

Key Characteristics

  • Research-Oriented: Intentionally developed to study and document the potential for AI misuse.
  • Restricted Access: Not intended for unrestricted public use due to its sensitive nature and potential for generating operationally relevant outputs related to criminal activity.
  • High-Risk Capabilities: Designed to explore and evaluate AI systems that may produce sensitive or potentially dangerous material.
  • Experimental System: Users are cautioned that the model may generate inaccurate, incomplete, misleading, offensive, dangerous, or legally incorrect information.

Authorized Use Cases

Access to JAX-XORTRON is restricted to individuals in legitimate professional, governmental, legal, academic, or AI-safety capacities. Authorized purposes include:

  • Research, evaluation, benchmarking, and red teaming.
  • Threat modeling, policy development, and legal analysis.
  • Education, defensive security, and criminal-justice research.
  • Forensic analysis or investigation of AI-enabled harm.

Users must certify their eligibility and agree not to use the model for committing, facilitating, directing, materially assisting, or concealing actual criminal activity or other unlawful conduct. Independent validation of model outputs and implementation of safeguards are the responsibility of the user.

Finetunes
2 models