M3Exam (Multimodal Memory) - Temporal Reasoning: leaderboard
M3Exam (2026) is a multimodal conversational-memory benchmark of 5,150 questions over long multi-session user-assistant conversations with images and documents; the frontier models here answer with the gold supporting sessions in context. This board is the LLM-judge score (0-100) on the temporal reasoning questions; higher is better.
Metric: LLM-judge score (0-100) on the temporal reasoning (TR) questions of 5,150 questions over 239 long multi-session user-assistant conversations with images and PDFs (15 personas); each model answers with the annotated gold supporting sessions and their images or rendered pages placed in context, no memory module; scored by a Qwen2.5-VL-32B-Instruct judge on a five-level rubric (0, 0.25, 0.5, 0.75, 1), printed 0-1 and x100 here; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Seed 2.0 Pro | 64.37 | #211 |
| 2 | Claude Opus 4.6 | 58.13 | #102 |
| 3 | Gemini 3.1 Pro (Preview) | 54.04 | #94 |
| 4 | Grok 4 | 51.5 | #307 |
| 5 | GLM-5.1 | 51.15 | #179 |
| 6 | Qwen 2.5 VL 7B Instruct | 50.63 | #1833 |
| 7 | Kimi K2.5 | 50.5 | #256 |
| 8 | GPT-5.4 | 49.5 | #123 |
| 9 | Qwen 3.6 Plus | 47.99 | #222 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=m3exam-multimodal-memory-temporal-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-13.