VideoGAIA - History: leaderboard
Metric: Pass@1 accuracy (%; the 44 history tasks; unified ReAct agent loop: 20 uniformly sampled frames plus web search, page visit and a thinking-with-videos tool that samples up to 20 more frames from a chosen segment, at most 40 steps; answers judged against the human-verified reference by GPT-5.5 (DeepSeek-V4-Pro fallback); mean of three runs). Source: arxiv.org. Saturation forecast: Around 2031. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 | 65.91 |
| 2 | Qwen 3.5 Plus | 63.64 |
| 3 | Claude Opus 4.6 | 59.09 |
| 4 | Kimi K2.6 | 59.09 |
| 5 | Qwen 3.7 Plus | 56.82 |
| 6 | Gemini 3.1 Pro (Preview) | 54.55 |
| 7 | GPT-5.5 | 54.55 |
| 8 | GPT-5.2 | 54.55 |
| 9 | Qwen 3.6 Plus | 54.55 |
| 10 | Seed 2.0 Pro | 54.55 |
| 11 | Kimi K3 | 52.27 |
| 12 | MiMo-V2.5 | 52.27 |
| 13 | GLM-5V Turbo | 52.27 |
| 14 | Kimi K2.5 | 50 |
| 15 | Qwen 3.8 Max | 50 |
Interactive version: theaggregate.ai/benchmark?slug=videogaia-history · How It Works · Data refreshed daily, snapshot 2026-09-26.