ChessImageBench — leaderboard
Metric: All Correct (%). Source: github.com. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Human Expert | 100 |
| 2 | Gemini 3.1 Pro (Preview) (Low) | 20.8 |
| 3 | Gemini 2.5 Pro | 7.2 |
| 4 | GPT-5.4 (Low) | 6.7 |
| 5 | GPT-5 | 4.8 |
| 6 | Grok 4 | 4.3 |
Interactive version: theaggregate.ai/benchmark?slug=chessimagebench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.