Long-horizon Agentic Reasoning on GAIA
70.9Pass@3AgentCPM-Explore + Long-context RL
Evaluation results
| Method | Links | |
|---|---|---|
| AgentCPM-Explore + Long-context RLTraining Steps=50 steps, Data Recipe=Long-context data recipe2026.06 | 70.9 | |
| AgentCPM-Explore + Long-context RLTraining Steps=25 steps, Data Recipe=Long-context data recipe2026.06 | 67.7 | |
| AgentCPM-ExploreData Recipe=Base Agent-enhanced model2026.06 | 66.1 |