MemEye (Full Context) - Multiple Choice: leaderboard
Long-term multimodal agent memory: questions about eight life-scenario conversation histories that require keeping fine-grained visual evidence and tracking visual states that change across sessions; MemEye (Guo et al., 2026). The model reads the whole native multimodal history (a 128k-token budget, oldest turns dropped first) with no memory system; exact-match accuracy in percent on 371 four-option questions, averaged over four answer rotations and then over the benchmark's 12 visual-granularity by memory-reasoning cells. Higher is better.
Metric: Exact-match accuracy (%; 371 four-option questions, averaged over four answer rotations and then macro-averaged over the 12 visual-granularity x memory-reasoning cells; full native multimodal history in context, 128k-token budget with oldest-first truncation, no memory system). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 Mini | 60.38 | #389 |
| 2 | Gemini 2.5 Flash Lite | 57.82 | #958 |
| 3 | GPT-4.1 Nano | 41.13 | #1992 |
| 4 | Qwen 3 VL 8B Instruct | 29.74 | #982 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=memeye-full-context-multiple-choice · How It Works · Data refreshed daily, snapshot 2026-11-12.