Video SimpleQA: leaderboard
Video SimpleQA evaluates factual grounding in large video language models with short-form, multi-hop, temporally grounded video questions.
Metric: F-score (%). Source: github.com. Source: benchmarklist.com. Saturation forecast: Around January 2027. 38 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 | 66.3 | #124 |
| 2 | Gemini 2.5 Pro | 62.6 | #152 |
| 3 | Gemini 2.5 Flash | 57 | #245 |
| 4 | GPT-4.5 | 54.1 | #274 |
| 5 | O4 Mini | 54 | #180 |
| 6 | GPT-4o | 49.3 | #343 |
| 7 | Claude 3.7 Sonnet | 36.2 | #249 |
| 8 | Claude Sonnet 4 | 35.6 | #199 |
| 9 | Qwen 2 VL 72B | 34.2 | #460 |
| 10 | Claude 3.5 Sonnet | 34 | #350 |
| 11 | InternVL3-78B | 33.8 | #352 |
| 12 | InternVL3-38B | 31.5 | #405 |
| 13 | Qwen 2.5 VL 7B | 25.3 | #643 |
| 14 | InternVL3-14B | 25.2 | #505 |
| 15 | InternVL3-8B | 23.5 | #620 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=video-simpleqa · How It Works · Data refreshed daily, snapshot 2026-10-20.