KOFFVQA - Table Understanding: leaderboard
KOFFVQA subscore for interpreting table images and structured visual information.
Metric: Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 82 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 03-25) | 96.67 | #123 |
| 2 | Gemini 2.0 Flash (001) | 96.67 | #360 |
| 3 | Gemini 2.5 Pro (Preview 05-06) | 96.67 | #147 |
| 4 | Gemini 2.5 Flash (Preview 05-20) | 96.67 | #216 |
| 5 | Gemini 2.5 Flash (Preview 04-17) | 95.67 | #237 |
| 6 | O4 Mini (2025-04-16) | 95.33 | #179 |
| 7 | O3 (2025-04-16) | 94 | #121 |
| 8 | GPT-4.5 (Preview) | 93.33 | #244 |
| 9 | Gemini 2.0 Flash (Preview) | 90 | #370 |
| 10 | Gemini 2.0 Flash Lite (001) | 90 | #544 |
| 11 | GPT-4.1 | 88.67 | #248 |
| 12 | O1 (2024-12-17) | 86.67 | #149 |
| 13 | Llama 4 Scout Instruct | 85.33 | #539 |
| 14 | GPT-4.1 Mini | 85 | #355 |
| 15 | Qwen 2.5 VL 72B Instruct | 84 | #377 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=koffvqa-table-understanding · How It Works · Data refreshed daily, snapshot 2026-10-20.