VideoMemBench - 32 Frames - Overall: leaderboard
Video-grounded multi-turn dialogue memory under sequential self-rollout, with model answers retained in history. Turn accuracy, episodic dependency accuracy and answer-chain accuracy are separate, overlapping views; the overall score is their equal mean. 326 dialogues; sequential self-rollout with full previous model answers; uniform timestamped video frames; no audio/subtitles; complete conversation history; exact deployment checkpoints and thinking modes unreported; native MC extraction and GPT-5.4 mini open-answer grading; percentages; 32 frames. Metric: Accuracy (%).
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 235B A22B (Benchmark Sweep Run: source=VideoMemBench; frames=32; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 29.5 | |
| 2 | Qwen 3.5 122B A10B (Benchmark Sweep Run: source=VideoMemBench; frames=32; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 27 | |
| 3 | Qwen 3.5 35B A3B (Benchmark Sweep Run: source=VideoMemBench; frames=32; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 27 | |
| 4 | Qwen 3.5 27B (Benchmark Sweep Run: source=VideoMemBench; frames=32; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 26.9 | |
| 5 | Qwen 3 VL 32B (Benchmark Sweep Run: source=VideoMemBench; frames=32; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 26.6 | |
| 6 | Qwen 3 VL 30B A3B (Benchmark Sweep Run: source=VideoMemBench; frames=32; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 24.2 | |
| 7 | Qwen 3.5 9B (Benchmark Sweep Run: source=VideoMemBench; frames=32; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 22.9 | |
| 8 | Qwen 3 VL 8B (Benchmark Sweep Run: source=VideoMemBench; frames=32; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 22.7 | |
| 9 | Qwen 3 VL 4B (Benchmark Sweep Run: source=VideoMemBench; frames=32; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 22.3 | |
| 10 | Qwen 2.5 VL 7B (Benchmark Sweep Run: source=VideoMemBench; frames=32; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 17.3 |
Interactive version: theaggregate.ai/benchmark?slug=videomembench-32-frames-overall · How It Works · Data refreshed daily, snapshot 2026-11-06.