VISTA (Video Grounding): leaderboard
VISTA (2026) tests open-set spatio-temporal video grounding: from a trimmed video and a natural-language query, a vision-language model must box the queried subject in every frame, over about 12,000 video-query pairs from six re-annotated datasets. Scored by mean spatio-temporal IoU in percent, pooled over referral and freeform queries; higher is better.
Metric: Mean spatio-temporal IoU (m_vIoU, %): per-frame box IoU summed over the frames where predicted and true tubes overlap in time, divided by their temporal union, pooled over referral and freeform queries (R&F), over the 11,814 video-query pairs of VISTA (HCSTVG-v1 and v2, VidVRD, VidSTG, MeViS and RVOS videos re-annotated with an interaction taxonomy), zero-shot on sub-sampled frames; the model must localize the queried subject with a box in every frame; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 9 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 8B | 63.96 | #917 |
| 2 | VISTA CogVLM Grounding (checkpoint unspecified) | 54.7 | |
| 3 | InternVL2.5-8B | 49.73 | #2370 |
| 4 | VISTA Sphinx-v2 (checkpoint unspecified) | 45.82 | |
| 5 | MiniGPT-v2 | 45.78 | |
| 6 | Qwen-VL-Chat | 45.49 | #7860 |
| 7 | Shikra-7B | 31.21 | |
| 8 | LLaVA-Grounding-7B | 27.11 | |
| 9 | Ferret-7B | 20.53 |
Interactive version: theaggregate.ai/benchmark?slug=vista-video-grounding · How It Works · Data refreshed daily, snapshot 2026-11-21.