Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Report Generation on DeepResearch Bench 2025 (test)

49.5Comprehensiveness

Gemini DeepResearch

33.17237.41141.6545.889Feb 3, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
49.549.550.15049.7
2026.02
48.348.150.750.849.3
2026.02
46.543.749.447.246.5
2026.02
45.342.847.644.745
2026.02
45.245.849.247.346.8
2026.02
43.442.548.84845.2
2026.02
41.741.848.241.343.4
2026.02
41.541.84946.944.3
2026.02
41.139.748.546.543.4
2026.02
40.539.848.245.442.9
2026.02
40.540.948.245.743.4
2026.02
39.534.446.244.340.5
2026.02
39.435.44643.540.6
2026.02
38.938.547.344.541.8
2026.02
38.537.94643.441
2026.02
33.829.939.14035