VideoMemBench - 128 Frames - Turn 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; 128 frames. Metric: Accuracy (%).
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 25 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 (Benchmark Sweep Run: source=VideoMemBench; frames=128; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 52.3 | |
| 2 | Kimi K3 (Benchmark Sweep Run: source=VideoMemBench; frames=128; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 50.5 | |
| 3 | Gemini 3.8 Flash (Benchmark Sweep Run: source=VideoMemBench; frames=128; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 50.4 | |
| 4 | GPT-5.6 Sol (Benchmark Sweep Run: source=VideoMemBench; frames=128; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 50 | |
| 5 | Gemini 3.5 Flash (Benchmark Sweep Run: source=VideoMemBench; frames=128; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 49.6 | |
| 6 | GPT-5.5 (Benchmark Sweep Run: source=VideoMemBench; frames=128; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 48 | |
| 7 | Claude Sonnet 5 (Benchmark Sweep Run: source=VideoMemBench; frames=128; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 47.8 | |
| 8 | Gemini 3.1 Pro (Preview) (Benchmark Sweep Run: source=VideoMemBench; frames=128; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 45.7 | |
| 9 | GPT-5.4 (Benchmark Sweep Run: source=VideoMemBench; frames=128; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 44.4 | |
| 10 | Qwen 3 VL 235B A22B (Benchmark Sweep Run: source=VideoMemBench; frames=128; self-rollout; exact-checkpoint-and-thinking-mode=unreported) | 43.1 |
Interactive version: theaggregate.ai/benchmark?slug=videomembench-128-frames-turn-accuracy · How It Works · Data refreshed daily, snapshot 2026-11-06.