VSAS-Bench (Synchronous): leaderboard
VSAS-Bench (2026) tests vision-language models as streaming visual assistants on 92 videos with dense per-second annotations, under the synchronous protocol where the stream waits for the model. A GPT-5 judge scores each timestep's response against the reference; mean accuracy (%) over Present, Cumulative and Future tasks, higher is better.
Metric: Mean average accuracy (%) over all Present, Cumulative and Future tasks 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: Around February 2027. 12 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-5 | Low | 77.1 | #185 |
| 2 | Qwen 3 VL 32B Instruct | 59.5 | #607 | |
| 3 | InternVL3-38B | 55.8 | #948 | |
| 4 | Qwen 3 VL 8B Instruct | 54.3 | #1043 | |
| 5 | Qwen 3 VL 4B Instruct | 51.6 | #1348 | |
| 6 | Dispider | 46.8 | #3834 | |
| 7 | InternVL3-8B | 46.3 | #1795 | |
| 8 | Qwen 3 VL 2B Instruct | 42.4 | #2389 | |
| 9 | InternVL3-2B | 36.9 | #3318 | |
| 10 | StreamBridge [VSAS-Bench checkpoint unspecified] | 36.8 | ||
| 11 | Flash-VStream-7B | 13.5 | ||
| 12 | VideoLLM-Online | 11.3 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=vsas-bench-synchronous · How It Works · Data refreshed daily, snapshot 2026-11-22.