List your company

General AI Assistant tasks on GAIA level2 Text-only

57.14AccuracyTodoEvolve
12.57624.145535.71547.2845Feb 8, 2026

Evaluation results

MethodLinks
TodoEvolveBackbone=Kimi-K22026.0257.140.0282323.659.1
JoyAgentBackbone=Kimi-K22026.02300.0027304.384.1
FlowSearchBackbone=Kimi-K22026.0225.710.0069159.144.6
Flash-SearcherBackbone=Kimi-K22026.0225.710.0255305.678
OWLBackbone=Kimi-K22026.0224.290.013299.786.2
Co-SightBackbone=Kimi-K22026.0217.140.0018250.232.6
OAgentsBackbone=Kimi-K22026.0215.710.0317292.128.7
AgentOrchestraBackbone=Kimi-K22026.0214.290.0149222.757.7