TOC-Bench - Hallucination Diagnostic Accuracy: leaderboard
Metric: Hallucination diagnostic accuracy (%, HDA): accuracy on the nonexistent-subject and nonexistent-event question variants, where the model must reject a plausible but visually unsupported object or event, on TOC-Bench (2,323 human-verified questions over 1,951 videos from Charades, Perception Test, OVIS and MOSE, grounded in per-frame object tracks and kept only when not answerable from text alone, single frames or shuffled frames), mostly 32 uniformly sampled frames, deterministic scoring of multiple-choice, statement-pair, numeric and ordering formats; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 23 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Grok 4.3 | 53.8 |
| 2 | Kimi K2.6 | 52.9 |
| 3 | Gemini 3.1 Pro (Preview) | 47.7 |
| 4 | Qwen 3 VL 8B (Thinking) | 43.8 |
| 5 | GPT-5.5 | 42.8 |
| 6 | Gemini 3.1 Flash Lite (Preview) | 41.2 |
| 7 | Seed 2.0 Lite | 41.2 |
| 8 | Qwen 2.5 VL 7B | 35.3 |
| 9 | MiMo-V2-Omni | 35.1 |
| 10 | GLM-5V Turbo | 34.7 |
| 11 | GPT-5.4 Mini | 28.4 |
| 12 | InternVL3-8B | 27.8 |
| 13 | MiniCPM-V-2.6 | 22.3 |
Interactive version: theaggregate.ai/benchmark?slug=toc-bench-hallucination-diagnostic-accuracy · How It Works · Data refreshed daily, snapshot 2026-10-07.