VSAS-Bench (Synchronous) - Future: leaderboard
Metric: Mean average accuracy (%) on Future tasks (predicting upcoming events from current cues) under the synchronous protocol (the model runs in lockstep with the stream, so inference latency does not count; video VLMs adapted with the sliding-window memory policy), over VSAS-Bench's 92 videos with dense 1 FPS free-form annotations; each timestep's response is judged against the annotation by GPT-5 (medium reasoning) on a 0-1 scale, shown as %; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 (High) | 85.1 |
| 2 | GPT-5 (Medium) | 84.4 |
| 3 | GPT-5 (Low) | 81.8 |
| 4 | Qwen 3 VL 8B Instruct | 35.5 |
| 5 | InternVL3-38B | 34.7 |
| 6 | Qwen 3 VL 32B Instruct | 31.4 |
| 7 | Qwen 3 VL 4B Instruct | 19.5 |
| 8 | InternVL3-8B | 14.6 |
Interactive version: theaggregate.ai/benchmark?slug=vsas-bench-synchronous-future · How It Works · Data refreshed daily, snapshot 2026-10-07.