VideoDR: leaderboard
Video deep research benchmark for video-conditioned open-domain QA requiring cross-frame visual anchors, web retrieval, and multi-hop reasoning.
Metric: Accuracy (%). Source: videodr-benchmark.github.io. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 69 | #110 |
| 2 | Gemini 3 Pro (Preview) | 69 | #66 |
| 3 | GPT-4o | 42 | #343 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=videodr · How It Works · Data refreshed daily, snapshot 2026-10-20.