Extreme-long-horizon agent tasks require multi-hour runtimes spanning hundreds of tool interactions and consume between 700,000 and 1,000,000 tokens per rollout across context windows.

From the articleAlibaba Qwen Introduces QwenGyre: Elastic Reinforcement Learning RL for XLong-Horizon AgentsAlso in Alibaba details QwenGyre to accelerate extreme-horizon agent training