Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

General AI Assistant Reasoning research benchmarks · SOTA2 Research

General AI Assistant Reasoning

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

GAIA-Text-103 1.0 (test)

Claude-3.7-Sonnet

62.1Overall Accuracy

26

Apr 23, 2026

BrowseComp-zh (BC-zh)

OPENAI DEEPRESEARCH

42.9Pass@1 Accuracy

19

Mar 10, 2026

60.12Accuracy

18

Feb 26, 2026

GAIA (File/Reasoning/Others)

56.21Accuracy

18

Feb 26, 2026

63.33Accuracy

18

Feb 26, 2026

OPENAI DEEPRESEARCH

67.4Pass@1 Accuracy

17

Mar 10, 2026

BrowseComp (BC)

OPENAI DEEPRESEARCH

51.5Pass@1 Accuracy

17

Mar 10, 2026

AgenticQwen-30B-A3B

344.1Average Latency (s)

3

Apr 24, 2026

43.7Pass@1

3

Feb 26, 2026

Popular datasets

GAIA BrowseComp GAIA-Text-103

Follow for updates

Join SOTA2 Contact