MEGA-Bench Task - Road Map Find Highway Between Two Place: leaderboard
The "Road Map Find Highway Between Two Place" subtask of MEGA-Bench Task. A single task from MEGA-Bench, a large-scale multimodal evaluation suite that scores vision-language models across 500+ real-world tasks spanning perception, reasoning, OCR, charts, documents, and multi-image understanding.
Metric: Task Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 44 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude 3.5 Sonnet (20240620) | 82.4 | #299 |
| 2 | Gemini 2.0 Flash (Preview) | 82.4 | #370 |
| 3 | Claude 3.5 Sonnet (20241022) | 76.5 | #291 |
| 4 | Gemini 2.5 Pro (Preview 03-25) | 76.5 | #123 |
| 5 | GPT-4o | 64.7 | #343 |
| 6 | GPT-4o Mini | 64.7 | #599 |
| 7 | Gemini 1.5 Pro (002) | 64.7 | #349 |
| 8 | Gemini 1.5 Flash (002) | 64.7 | #583 |
| 9 | InternVL2.5-78B | 64.7 | #431 |
| 10 | GLM-4.1V-9B (Thinking) | 64.7 | #472 (GLM-4.1V-9B) |
| 11 | InternVL3-78B | 52.9 | #352 |
| 12 | Qwen 2 VL 72B | 52.9 | #460 |
| 13 | Pixtral-12B | 52.9 | #808 |
| 14 | Llama 3.2 11B | 52.9 | #1199 |
| 15 | Gemma 3 27B (IT) | 41.2 | #523 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-road-map-find-highway-between-two-place · How It Works · Data refreshed daily, snapshot 2026-10-20.