InfiR2: A Comprehensive FP8 Training Recipe for Reasoning-Enhanced Language Models
Published in Under Review, 2025
This manuscript presents an FP8 training recipe for reasoning-enhanced language models, with attention to training stability and deployment-oriented precision consistency.
Recommended citation: Wenjun Wang, Shuo Cai, Congkai Xie, Mingfa Feng, Yiming Zhang, Zhen Li, Kejing Yang, Ming Li, Jiannong Cao, Yuan Xie, Hongxia Yang. InfiR2: A Comprehensive FP8 Training Recipe for Reasoning-Enhanced Language Models. Under review, 2025.
Download Paper