List your company

Multi-tool calling on VitaBench (held-out test)

63.1Success RateSkill-MAS-optimized
-0.048816.345632.7449.1344Jun 17, 2026

Evaluation results

MethodLinks
Skill-MAS-optimizedMeta-agent=DeepSeek-V4-Flash2026.0663.1
EvoAgentMeta-agent=DeepSeek-V4-Flash2026.0661.9
AOrchestraMeta-agent=DeepSeek-V4-Flash2026.0658.33
AFlowMeta-agent=DeepSeek-V4-Flash2026.0655.95
Skill-MAS-initMeta-agent=DeepSeek-V4-Flash2026.0653.57
Skill-MAS-optimizedMeta-agent=Qwen3.5-Plus2026.0648.8
MAS-OrchestraMeta-agent=DeepSeek-V4-Flash2026.0647.62
AFlowMeta-agent=Qwen3.5-Plus2026.0645.24
AOrchestraMeta-agent=Qwen3.5-Plus2026.0640.48
Skill-MAS-initMeta-agent=Qwen3.5-Plus2026.0636.9
MAS-OrchestraMeta-agent=Qwen3.5-Plus2026.0635.71
EvoAgentMeta-agent=Qwen3.5-Plus2026.0634.52
MAS2Meta-agent=DeepSeek-V4-Flash2026.0633.33
Skill-MAS-optimizedMeta-agent=Gemini-3.1-Flash2026.0628.6
MAS2Meta-agent=Qwen3.5-Plus2026.0622.62
AFlowMeta-agent=Gemini-3.1-Flash2026.0619.05
Skill-MAS-initMeta-agent=Gemini-3.1-Flash2026.0616.67
Skill-MAS-optimizedMeta-agent=GPT-5.4-Nano2026.0615.48
AOrchestraMeta-agent=Gemini-3.1-Flash2026.0614.29
EvoAgentMeta-agent=Gemini-3.1-Flash2026.0610.71
MAS-OrchestraMeta-agent=Gemini-3.1-Flash2026.069.52
MAS-OrchestraMeta-agent=GPT-5.4-Nano2026.069.52
MAS2Meta-agent=Gemini-3.1-Flash2026.068.33
EvoAgentMeta-agent=GPT-5.4-Nano2026.068.33
Skill-MAS-initMeta-agent=GPT-5.4-Nano2026.065.95
AOrchestraMeta-agent=GPT-5.4-Nano2026.064.76
AFlowMeta-agent=GPT-5.4-Nano2026.062.38
MAS2Meta-agent=GPT-5.4-Nano2026.062.38