List your company

Agent tool-use planning on GeoScenario 9 Tools 116 (test)

73.8TSAQwen3-8B
33.2443.7754.364.83May 6, 2026

Evaluation results

MethodLinks
Qwen3-8BApproach=Meta2026.0573.873.553.130.287.983.561.7
Qwen3-235BApproach=Meta2026.0568.568.554.735.696.691.776
Qwen2.5-72BApproach=Meta2026.0566.866.349.136.59792.352.2
Qwen3-235BApproach=Base2026.0546.145.937.939.972.26948.6
Qwen2.5-72BApproach=Base2026.0541.640.431.539.165.556.420.4
Qwen3-8BApproach=Base2026.0534.834.726.538.35648.721.4