GST-Bench - Global Position (Visual): leaderboard
Metric: Mean accuracy (%; predicted top-down map position of a previously seen object within 100 to 300 pixels, averaged over five thresholds; target marked by a bounding box in the video; 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 | 42.23 |
| 2 | GPT-5 | 34.28 |
| 3 | Gemini 2.5 Pro | 31.8 |
| 4 | Seed 1.8 | 27.96 |
| 5 | GPT-4o | 22.36 |
| 6 | InternVL3.5-8B | 19.06 |
| 7 | Qwen 3 VL 32B | 18.05 |
Interactive version: theaggregate.ai/benchmark?slug=gst-bench-global-position-visual · How It Works · Data refreshed daily, snapshot 2026-09-29.