CountQA: leaderboard
Visual counting question-answering benchmark reported in Qwen's Qwen3.7-Plus comparison table.
Metric: Score (%). Source: qwen.ai. Source: benchmarklist.com. Saturation forecast: Around 2029. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.7 Plus | 77 | #92 |
| 2 | Gemini 3.1 Pro (Preview) | 72.8 | #58 |
| 3 | Qwen 3.6 Plus | 71.7 | #120 |
| 4 | GPT-5.4 (xHigh) | 58.4 | #79 (GPT-5.4) |
| 5 | Claude Opus 4.6 (Max) | 32.5 | #63 (Claude Opus 4.6) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=countqa · How It Works · Data refreshed daily, snapshot 2026-10-20.