EMem-Bench - Passive Observation: leaderboard
Metric: Success rate (%; the 1036 Passive Observation episodes, finding an object seen earlier in an object-dense scene; full-context condition: the complete interaction history as sanitized text plus the current 500 x 500 RGB view, one primitive action per call with deterministic decoding and at most eight target-task steps in AI2-THOR and ProcTHOR homes; an episode succeeds only if the simulator reaches the target terminal state). Source: arxiv.org. Saturation forecast: Around December 2026. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash | 67 |
| 2 | Gemini 2.5 Pro | 63.5 |
| 3 | Qwen 3.6 27B | 61 |
| 4 | GPT-5.4 | 60.4 |
| 5 | GPT-5.4 Mini | 55.1 |
| 6 | InternVL3-38B | 47.9 |
| 7 | Qwen 3 VL 32B Instruct | 47.1 |
| 8 | Qwen 3 VL 4B Instruct | 46.4 |
| 9 | Qwen 3 VL 8B Instruct | 41.4 |
| 10 | InternVL3-8B | 20.5 |
Interactive version: theaggregate.ai/benchmark?slug=emem-bench-passive-observation · How It Works · Data refreshed daily, snapshot 2026-09-26.