GameHorizon - Offline Overall: leaderboard

Metric: Mean accuracy over T1-T3 (%). Source: gamehorizon-suite.github.io. Saturation forecast: Around June 2027. 43 models tracked. Updated 18 Nov 2026.

Top models

#ModelScoreOverall rank
1GPT-6 Astra80.2#7
2Gemini 3.8 Flash77.3#39
3Gemini 3.7 Flash76.7#38
4Gemini 3.6 Flash75.3#77
5GPT-5.6 Sol74.8#33
6Kimi K374.5#60
7Gemini 3.5 Flash74.4#106
8GPT-5.573.2#50
9Gemini 3.1 Pro (Preview)72.8#99
10Seed 2.1 Turbo72.7#165
11Seed 2.1 Pro72.1#142
12Seed 2.0 Pro71.4#218
13Claude Fable 571.2#29
14GPT-5.6 Terra70.7#69
15Qwen 3.7 Plus68.7#215

No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=gamehorizon-offline-overall · How It Works · Data refreshed daily, snapshot 2026-11-18.