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

Long-form Deep Research Report Generation Benchmarks

51.9DeepResearchBench II Score

Claude-4.6-Sonnet-Thinking

26.83633.34339.8546.357Jun 13, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
51.946.760.553
2026.06
484659.951.3
45.44759.750.5
2026.06
45.249.459.751.4
2026.06
44.845.560.950.4
2026.06
44.546.963.451.6
2026.06
44.245.758.649.5
2026.06
42.845.653.547.3
4248.961.551.1
2026.06
41.94662.550.1
2026.06
41.746.558.748.7
40.742.65144.8
39.653.350.747.9
2026.06
29.9---
2026.06
29.738.443.937.3
2026.06
27.83641.735.2
2026.06
--61.4-