GameHorizon - Offline T1 Single-Horizon Action: leaderboard
Metric: Accuracy (%). Source: gamehorizon-suite.github.io. Saturation forecast: Around 2029. 43 models tracked. Updated 18 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 69.4 | #7 |
| 2 | Gemini 3.7 Flash | 66.2 | #38 |
| 3 | Gemini 3.8 Flash | 65.9 | #39 |
| 4 | GPT-5.6 Sol | 65.3 | #33 |
| 5 | Gemini 3.5 Flash | 64.9 | #106 |
| 6 | GPT-5.5 | 64.6 | #50 |
| 7 | Gemini 3.1 Pro (Preview) | 64.5 | #99 |
| 8 | Kimi K3 | 64.4 | #60 |
| 9 | Seed 2.1 Turbo | 63.1 | #165 |
| 10 | Claude Fable 5 | 62.9 | #29 |
| 11 | Seed 2.1 Pro | 62.9 | #142 |
| 12 | Gemini 3.6 Flash | 62.3 | #77 |
| 13 | GPT-5.6 Terra | 62.1 | #69 |
| 14 | GPT-5.6 Luna | 61.6 | #129 |
| 15 | Kimi K2.6 | 61.2 | #188 |
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-t1-single-horizon-action · How It Works · Data refreshed daily, snapshot 2026-11-18.