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

Multi-turn task completion on TelepathyGym

90.2Task Completion Score

Gemini-1.5-Pro

1.48824.51947.5570.581Apr 8, 2026
Updated 3mo ago

Evaluation Results

MethodLinks
2026.04
90.2
2026.04
85.4
2026.04
63.4
2026.04
63.4
2026.04
63.2
2026.04
58.5
2026.04
56.1
2026.04
56.1
2026.04
55.8
2026.04
52.4
2026.04
51
2026.04
48.8
2026.04
4.9