AI Decathlon - Overall: leaderboard

AI Decathlon Timed Challenge (Overall) evaluating coding agents on 60-100 minute engineering, math, spatial, and interactive design tasks.

Metric: Weighted Score (0-100 across 13 timed tasks). Source: octathlon.wendystars.com. Saturation forecast: Around March 2027. 25 models tracked. Updated 22 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6.2 AstraxHigh84#4
2GPT-6.2 SolxHigh74.3#10
3Gemini 4 NeonxHigh70.6#3
4GPT-6 AstraxHigh70.3#11
5Claude Fable 5.2xHigh68.5#2
6GPT-6.1 SolxHigh67.8#15
7Claude Opus 5.5xHigh62.9#9
8Gemini 4 CarbonHigh59.4#7
9Claude Sonnet 5.5xHigh57.3#24
10Gemini 4 HydrogenxHigh56.6#16
11GPT-6.1 TerraxHigh45#25
12Claude Haiku 5.5xHigh40.7#75
13Kimi K3.1 ProMax37.7#35
14DeepSeek V4.1 FlashMax34.7#69
15GPT-5.6 SolxHigh33.4#36

Interactive version: theaggregate.ai/benchmark?slug=ai-decathlon-overall · How It Works · Data refreshed daily, snapshot 2026-11-22.