zenlm/zen-guard-gen-8b

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-8b is an 8 billion parameter generative content-safety and guardrail model. It is a repackaged version of Qwen/Qwen3Guard-Gen-8B by Alibaba Qwen, designed to provide permissively-licensed content moderation capabilities. This model is specifically engineered to act as a guardrail, filtering and ensuring the safety of generated content.

Loading preview...

Overview

zenlm/zen-guard-gen-8b is an 8 billion parameter generative content-safety and guardrail model. It is a repackaged version of the Qwen/Qwen3Guard-Gen-8B model developed by Alibaba Qwen, distributed under an Apache-2.0 license. This model is part of the OSS-clean Zen model line, providing a permissively-licensed option for content moderation.

Key Capabilities

  • Generative Content Safety: Functions as a guardrail to ensure the safety and appropriateness of generated text.
  • 8 Billion Parameters: A dense 8B parameter model, offering robust performance for its intended task.
  • Qwen3 Architecture: Built on the Qwen3ForCausalLM architecture, leveraging the capabilities of the Qwen3Guard-Gen family.
  • Permissively Licensed: Available under the Apache-2.0 license, making it suitable for a wide range of commercial and research applications.

Good For

  • Implementing content moderation and safety filters in applications.
  • Developing guardrail systems for other large language models.
  • Use cases requiring a permissively-licensed generative safety model.

Note: This model has been superseded by zenlm/zen3-guard, which is the canonical name for the updated version. The weights for zen-guard-gen-8b remain available for reproducibility.

Finetunes
1 models