List your company

Multi-hop dynamic question answering on BrowseComp-Plus (test)

27.38Accuracy (%)Skill-MAS-optimized
4.021610.085816.1522.2142Jun 17, 2026

Evaluation results

MethodLinks
Skill-MAS-optimizedMeta-agent=GPT-5.4-Nano2026.0627.38
EvoAgentMeta-agent=GPT-5.4-Nano2026.0623.81
Skill-MAS-optimizedMeta-agent=Qwen3.5-Plus2026.0623.81
Skill-MAS-optimizedMeta-agent=Gemini-3.1-Flash2026.0623.21
AOrchestraMeta-agent=GPT-5.4-Nano2026.0622.62
Skill-MAS-optimizedMeta-agent=DeepSeek-V4-Flash2026.0622.62
Skill-MAS-initMeta-agent=Qwen3.5-Plus2026.0620.24
Skill-MAS-initMeta-agent=GPT-5.4-Nano2026.0619.64
Skill-MAS-initMeta-agent=Gemini-3.1-Flash2026.0619.05
AOrchestraMeta-agent=Qwen3.5-Plus2026.0619.05
AFlowMeta-agent=GPT-5.4-Nano2026.0618.45
AOrchestraMeta-agent=DeepSeek-V4-Flash2026.0618.45
AFlowMeta-agent=Gemini-3.1-Flash2026.0617.86
EvoAgentMeta-agent=DeepSeek-V4-Flash2026.0616.67
EvoAgentMeta-agent=Gemini-3.1-Flash2026.0616.07
MAS-OrchestraMeta-agent=DeepSeek-V4-Flash2026.0615.72
EvoAgentMeta-agent=Qwen3.5-Plus2026.0615.48
AFlowMeta-agent=DeepSeek-V4-Flash2026.0615.48
Skill-MAS-initMeta-agent=DeepSeek-V4-Flash2026.0615.48
AOrchestraMeta-agent=Gemini-3.1-Flash2026.0614.88
AFlowMeta-agent=Qwen3.5-Plus2026.0612.5
MAS2Meta-agent=Qwen3.5-Plus2026.0611.52
MAS2Meta-agent=Gemini-3.1-Flash2026.0611.39
MAS2Meta-agent=GPT-5.4-Nano2026.0610.71
MAS-OrchestraMeta-agent=Qwen3.5-Plus2026.0610.12
MAS-OrchestraMeta-agent=Gemini-3.1-Flash2026.068.9
MAS-OrchestraMeta-agent=GPT-5.4-Nano2026.067.74
MAS2Meta-agent=DeepSeek-V4-Flash2026.064.92