EgoPoint-Bench - Simulation Spatial Context: 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 Spatial Context questions (location and reachability) among the 1,929 simulation test questions. Accuracy in percent; higher is better.

Metric: Accuracy (%) on the Spatial Context questions among the 1,929 held-out simulation test questions (HM3D, HSSD and AI2-THOR scenes); 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 May 2028. 12 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro70.47#146
2Qwen 3 VL 8B Instruct70.34#1043
3GPT-5 Mini67.65#373
4Gemini 3 Flash66.58#180
5GPT-5.2 Instant66.14#309
6GLM-4.6V-Flash66.14#1049
7Qwen 3 VL 32B Instruct65.64#607
8InternVL3.5-8B-Instruct63.51#1911
9InternVL3.5-14B-Instruct62.07#1747
10InternVL3.5-2B-Instruct61.5#6576
11LLaVA-NeXT-7B55.92#5045
12LLaVA-v1.5-7B54.86#7642

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=egopoint-bench-simulation-spatial-context · How It Works · Data refreshed daily, snapshot 2026-11-22.