AI Model Leaderboard

Agentic work ranking, Model families, sorted by Overall
Rank Model Provider Weights Overall Coding Agents Reasoning Math Knowledge Cost
1 Gemini 4 Carbon New Google 66.8 68.2 65.2 70.6 66.7 66.9 $4.00
5 Gemini 4 Argon Google 64.7 64.5 63.7 68.9 64.9 66.1 $4.00
22 Gemini 3.8 Flash Google 60.9 59.5 60.2 65.9 58.8 62.0 $1.50
28 Gemini 3.7 Flash Google 60.4 59.6 59.8 64.4 59.4 60.9 $1.50
48 Gemini 3.6 Flash Google 58.6 57.4 58.5 60.9 56.9 58.5 $1.50
56 Gemini 3.5 Flash Google 58.0 57.0 57.3 62.4 57.4 59.5 $3.38
62 Gemini 3.1 Pro Preview Google 57.0 54.9 56.1 63.9 57.5 62.3 $4.50
77 Gemini 3 Flash Preview Google 55.1 53.9 54.9 58.1 55.5 57.0 $1.13
78 Gemini 3 Pro Preview Google 55.1 55.4 53.9 59.1 55.8 58.9 $4.50
94 Gemini 3.5 Flash Lite Google 52.5 52.0 52.8 51.8 49.9 50.4 $0.85
116 Gemini 3.1 Flash Lite Preview Google 48.8 52.9 47.9 45.7 –No results 46.0 $0.56
117 Gemini 2.5 Pro Google 48.8 48.3 48.4 51.2 50.3 –No results $3.44
119 Gemini 2.5 Pro Preview 06-05 Google 48.4 50.4 47.7 48.1 49.3 51.9 $3.44
131 Gemini 2.5 Flash (Sep 2025) Google 45.9 50.9 42.4 51.9 –No results 48.8 –No price
133 Gemini 2.5 Flash Google 45.8 –No results 45.6 46.7 47.5 –No results $0.85
136 Gemini 3.1 Flash Lite Google 45.2 49.4 42.6 48.8 51.3 43.1 $0.56
157 Gemini 1.5 Pro (Sept 2024) Google 39.2 36.3 39.7 41.7 39.5 37.4 –No price
173 Gemini 1.5 Flash (Sep 2024) Google 34.6 34.3 34.4 36.1 36.8 27.0 –No price