EVID-Bench: leaderboard
Metric: Point-level accuracy (%) over all 222 manipulated videos: share of ground-truth misinformation points the model identifies, using the retrieval-augmented verification pipeline of the paper (chain-of-thought analysis, up to six YouTube search-verify-reflect rounds, 64 frames at 480p, temperature 0); each video has 3 to 5 annotated misinformation points, matched by a majority of three LLM judges; mean of three runs; higher is better. Source: arxiv.org. Saturation forecast: Around April 2028. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 61.43 |
| 2 | Claude Opus 4.6 | 58.34 |
| 3 | Gemini 3.1 Pro (Preview) | 55.9 |
| 4 | Claude Sonnet 4.6 | 55.13 |
| 5 | GPT-5.4 | 53.77 |
| 6 | Gemini 3 Flash | 53.6 |
| 7 | Qwen 3.5 Plus | 53.4 |
| 8 | Qwen 3 VL 235B A22B Instruct | 48.8 |
| 9 | GPT-5.4 Mini | 46.97 |
Interactive version: theaggregate.ai/benchmark?slug=evid-bench · How It Works · Data refreshed daily, snapshot 2026-09-29.