zenlm/zen-guard-gen

TEXT GENERATIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 17, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

zenlm/zen-guard-gen is an 8 billion parameter content-safety and guardrail model, rebased from Alibaba Qwen's Qwen3Guard-Gen-8B. It is designed for commercial use under an Apache-2.0 license, providing robust content moderation capabilities. This model is specifically engineered to filter and manage generated text, ensuring adherence to safety guidelines.

Loading preview...

zen-guard-gen: Content Safety and Guardrail Model

zenlm/zen-guard-gen is an 8 billion parameter language model specifically designed for content safety and guardrail applications. This model is a repackaged version of Alibaba Qwen's Qwen3Guard-Gen-8B, ensuring a clean Apache-2.0 license for commercial use. It was rebased from the prior Qwen2.5-3B base to facilitate broader commercial adoption.

Key Capabilities

  • Content Moderation: Functions as a robust guardrail to filter and manage generated text.
  • Safety Compliance: Helps ensure AI-generated content adheres to predefined safety guidelines.
  • Commercial Use: Licensed under Apache-2.0, making it suitable for commercial applications without restrictive research licenses.

Good For

  • Implementing safety layers in large language model applications.
  • Filtering potentially harmful or undesirable content from user inputs or model outputs.
  • Developers seeking an open-source, commercially viable guardrail solution.

Note: This model has been superseded by zenlm/zen3-guard, which is the canonical and recommended version for new projects. The weights for zen-guard-gen remain available for reproducibility purposes.

Finetunes
2 models