ResearchBenchmarksVision-Language Understanding on MMStar (test)Follow63.73AccuracyETTC40.319646.397352.47558.5527May 29, 2026Evaluation ResultsMethodMethodLinksAccuracyETTCAggregation Method=ETTCAggregation Method=ETTC2026.0563.73Qwen-72BModel Scale=72BModel Scale=72B2026.0562.56VotingAggregation Method=VotingAggregation Method=Voting2026.0561Qwen-7BModel Scale=7BModel Scale=7B2026.0556.77Qwen-32BModel Scale=32BModel Scale=32B2026.0556.34AverageAggregation Method=MeanAggregation Method=Mean2026.0554.22Qwen-3BModel Scale=3BModel Scale=3B2026.0541.22