MEGA-Bench Task - Google Street View Circle Reasoning — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 53.3 |
| 2 | Gemini 2.0 Flash (Preview) | 40 |
| 3 | MiniCPM-V-2.6 | 33.3 |
| 4 | Claude 3.5 Sonnet (20240620) | 33.3 |
| 5 | InternVL3-78B | 26.7 |
| 6 | Qwen 2 VL 7B | 26.7 |
| 7 | Qwen 2 VL 2B | 26.7 |
| 8 | Gemini 1.5 Pro (002) | 26.7 |
| 9 | GPT-4o | 20 |
| 10 | Gemma 3 27B (IT) | 20 |
| 11 | InternVL2-8B | 20 |
| 12 | Claude 3.5 Sonnet (20241022) | 20 |
| 13 | InternVL3-8B | 13.3 |
| 14 | GPT-4o Mini | 13.3 |
| 15 | InternVL2.5-78B | 13.3 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-google-street-view-circle-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.