GameHorizon - Online Thematic Tasks: leaderboard

Metric: Task success rate (%). Source: gamehorizon-suite.github.io. Saturation forecast: Around February 2027. 12 models tracked. Updated 18 Nov 2026.

Top models

#ModelScoreOverall rank
1GPT-6 Astra50#7
2Gemini 3.6 Flash30#77
3MiniMax-M310#277
4GPT-5.6 Luna10#129
5Qwen 3.5 397B A17B0#254
6Kimi K30#60
7Qwen 3.6 35B A3B0#480
8GPT-5.6 Terra0#69
9GLM-5V Turbo0#362
10uitars-1.5-7B0#7710
11InternVL3.5-8B0#1598
12Step3 VL 10B0#1084

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

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