EgoPoint-Bench - Real-World: leaderboard

EgoPoint-Bench (2026) asks multimodal models about the object a first-person hand points at, on held-out simulation scenes and real smart-glasses recordings, zero-shot; true/false and single-choice answers by exact match and open answers scored by a GPT-4o judge. This board covers the 1,162 real-world test questions across five capability dimensions. Accuracy in percent; higher is better.

Metric: Accuracy (%) over the 1,162 real-world smart-glasses test questions, all five dimensions; zero-shot (direct inference); true/false and single-choice questions by exact match, open-ended answers scored 0 to 1 by a GPT-4o judge; greedy decoding for open models, temperature 0 for API models; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 12 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro72#145
2Gemini 3 Flash71.84#175
3GPT-5.2 Instant66.76#304
4Qwen 3 VL 32B Instruct64.3#606
5GPT-5 Mini60.57#364
6InternVL3.5-14B-Instruct58.59#1736
7Qwen 3 VL 8B Instruct58.14#1029
8InternVL3.5-8B-Instruct57.09#1891
9GLM-4.6V-Flash56.42#1040
10InternVL3.5-2B-Instruct53.73#6517
11LLaVA-v1.5-7B47.19#7630
12LLaVA-NeXT-7B46.44#5032

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=egopoint-bench-real-world · How It Works · Data refreshed daily, snapshot 2026-11-21.