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

#ModelScore
1GPT-5.561.43
2Claude Opus 4.658.34
3Gemini 3.1 Pro (Preview)55.9
4Claude Sonnet 4.655.13
5GPT-5.453.77
6Gemini 3 Flash53.6
7Qwen 3.5 Plus53.4
8Qwen 3 VL 235B A22B Instruct48.8
9GPT-5.4 Mini46.97

Interactive version: theaggregate.ai/benchmark?slug=evid-bench · How It Works · Data refreshed daily, snapshot 2026-09-29.