ByteDance Seed
HomeModelsBlog & PublicationSeed EdgeJoin Us
HomeModelsBlog & PublicationSeed EdgeJoin Us
Try DolaTry in Playground

2026-07-06

EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments

Download PDF
PreviousNext

ABSTRACT

Pretraining scaling laws reveal that model capability improves predictably with data and compute. But learning from real world environments after deployment remains far less understood. Analyzing roughly 38,000 hours of agent interaction with the environment across 134 real world tasks, we find, to the best of our knowledge, the first evidence that overall performance during environment learning follows a log-sigmoid scaling law with remarkably high precision, reaching R^2 = 0.998. Across model generations, we also find that agent learning speed roughly doubles every three months. This discovery stems from EdgeBench, a suite of 134 real world tasks with ultra-long horizons, spanning scientific discovery, software engineering, combinatorial optimization, professional knowledge work, formal mathematics, and interactive games. Each task sustains at least 12 hours of continuous agent operation under rich, multilevel feedback, and is built through substantial expert effort. We publicly release 51 tasks and our full evaluation framework to accelerate the study of how agents learn from real world experience.

AUTHORS

Deyao Zhu, Xin Zhou, Shengling Qin, Xuekai Zhu, Hangliang Ding, Shu Zhong, Zixin Wen, Zhonglin Xie, Chenhui Gou, Linxuan Ren, Yueyang Wang, Junfeng Zhong, Rui Liu, Tian Gao, Yangguang Lin, Jingyuan Zhang, Maojia Song, Xuan Qi, Jinhong Wu, Chenyang Zhang, Yinzhu Piao, Ziru Niu, Hongbin Lin, Lingxiang Meng, Peng Tang, Chengyao Tang, Shanyu Wu, Huanyu Zheng, Yu Liu, Liya Zhu, He Wang, Ming Ding, Ziyu Wan, Hao Liu, Sibo Wang, Haotian Zhu, Xintian Zhang, Nan Chai, Yipeng Liu, Panhao Lai, Sihang Yuan, Zixin Su, Ge Zhang, Wangchunshu Zhou, Yantao Du, Wenhao Huang, Guang Shi

VENUE

arXiv

Advancing the frontier of intelligence, in service of humanity
Join ByteDance Seed
Models
Seed2.1
Seedance 2.5
Seedream 5.0 Pro
SeedRealtime
Seed Audio 1.0
Seed GR-RL
Learn More
Publications
Seed Edge
Seed STEM Fellows Program
Seed Campus Recruitment
Transparency
Copyright© 2026 Bytedance Seed
DisclaimerContact us : seed.feedback@bytedance.com
Seed2.1Seedance 2.5Seedream 5.0 ProSeedRealtimeSeed Audio 1.0Seed GR-RL
PublicationsSeed EdgeSeed STEM Fellows ProgramSeed Campus Recruitment
Transparency
Advancing the frontier of intelligence, in service of humanity
Join ByteDance Seed
Copyright© 2026 Bytedance Seed
Disclaimer
Contact us : seed.feedback@bytedance.com
Advancing the frontier of intelligence, in service of humanity
Contact us : seed.feedback@bytedance.com
Copyright© 2026 Bytedance SeedDisclaimer