We present GuardReasoner-Omni, a reasoning-based guardrail model designed to moderate text, image, video, and audio data. First, we construct a comprehensive training corpus comprising 181k samples spanning these four modalities. Our training pipeline follows a two-stage paradigm to incentivize the model to deliberate before making decisions: (1) conducting SFT to cold-start the model with explicit reasoning capabilities and structural adherence; and (2) performing RL with a concise correctness reward to preserve accurate reasoning while suppressing redundant generation. We release a suite of models scaled at 3B and 7B parameters. Extensive experiments demonstrate that GuardReasoner-Omni achieves superior performance compared to existing state-of-the-art baselines across various guardrail benchmarks.
@article{arxiv.2602.03328,
title = {GuardReasoner-Omni: A Reasoning-based Multi-modal Guardrail for Text, Image, Video, and Audio},
author = {Zhenhao Zhu and Yue Liu and Yanpei Guo and Wenjie Qu and Cancan Chen and Yufei He and Yibo Li and Yulin Chen and Tianyi Wu and Huiying Xu and Xinzhong Zhu and Jiaheng Zhang},
journal= {arXiv preprint arXiv:2602.03328},
year = {2026}
}