VideoMemBench - 64 Frames - Answer Chain Accuracy: 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; 64 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=64; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 15.6 | |
| 2 | Qwen 3.5 122B A10B (Benchmark Sweep Run: source=VideoMemBench; frames=64; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 15.2 | |
| 3 | Qwen 3.5 27B (Benchmark Sweep Run: source=VideoMemBench; frames=64; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 14.8 | |
| 4 | Qwen 3 VL 32B (Benchmark Sweep Run: source=VideoMemBench; frames=64; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 14.5 | |
| 5 | Qwen 3.5 35B A3B (Benchmark Sweep Run: source=VideoMemBench; frames=64; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 13.5 | |
| 6 | Qwen 3 VL 4B (Benchmark Sweep Run: source=VideoMemBench; frames=64; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 12.1 | |
| 7 | Qwen 3 VL 8B (Benchmark Sweep Run: source=VideoMemBench; frames=64; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 11.9 | |
| 8 | Qwen 3 VL 30B A3B (Benchmark Sweep Run: source=VideoMemBench; frames=64; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 11.5 | |
| 9 | Qwen 3.5 9B (Benchmark Sweep Run: source=VideoMemBench; frames=64; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 9.6 | |
| 10 | Qwen 2.5 VL 7B (Benchmark Sweep Run: source=VideoMemBench; frames=64; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 6.4 |
Interactive version: theaggregate.ai/benchmark?slug=videomembench-64-frames-answer-chain-accuracy · How It Works · Data refreshed daily, snapshot 2026-11-06.