EgoEsportsQA: leaderboard
Metric: Overall accuracy (%) on the 1,745 EgoEsportsQA multiple-choice questions from first-person Counter-Strike 2, Valorant and Overwatch 2 match clips (perception and reasoning levels), 1 fps sampling; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 71.58 |
| 2 | Seed 1.8 | 61.38 |
| 3 | Claude Sonnet 4.5 | 51.81 |
| 4 | Qwen 3 VL 8B Instruct | 50.26 |
| 5 | Gemini 3 Flash | 49.74 |
Interactive version: theaggregate.ai/benchmark?slug=egoesportsqa · How It Works · Data refreshed daily, snapshot 2026-10-07.