GST-Bench - Top-Down Selection (Hard): leaderboard
Metric: Accuracy (%; multiple choice of the top-down image of the explored scene, hard distractors; zero-shot, greedy decoding, official prompt templates, over human-verified questions on 6,790 minutes of simulator-rendered egocentric exploration video). Source: arxiv.org. Saturation forecast: Around December 2027. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 52.25 |
| 2 | GPT-4o | 45.51 |
| 3 | GPT-5 | 43.26 |
| 4 | Gemini 2.5 Pro | 41.01 |
| 5 | Seed 1.8 | 37.64 |
| 6 | Qwen 3 VL 32B | 32.58 |
| 7 | InternVL3.5-8B | 25.84 |
Interactive version: theaggregate.ai/benchmark?slug=gst-bench-top-down-selection-hard · How It Works · Data refreshed daily, snapshot 2026-09-29.