GeoCode Leaderboard — leaderboard
Leaderboard for large language models on geospatial code generation using AutoGEEval and AutoGEEval++ pass@k metrics on Google Earth Engine tasks.
Metric: AutoGEEval++ Pass@1 (self-reported). Source: benchmarklist.com. Status: saturation imminent. 28 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 | 70.93 |
| 2 | Claude 3.7 Sonnet | 70.35 |
| 3 | DeepSeek V3 (0324) | 70.25 |
| 4 | Qwen 2.5 Coder 32B | 66.77 |
| 5 | GPT-4.1 Mini | 66.56 |
| 6 | Qwen 2.5 32B | 65.45 |
| 7 | QwQ-32B | 60.69 |
| 8 | Qwen 3 32B | 60.67 |
| 9 | DeepSeek R1 | 60.42 |
| 10 | GPT-4o | 59.02 |
| 11 | O3 Mini (Medium) | 56.98 |
| 12 | Qwen 3 32B (Thinking) | 56.31 |
| 13 | GPT-4o Mini | 55.02 |
| 14 | Qwen 2.5 Coder 7B | 50.51 |
| 15 | Qwen 2.5 7B | 49.58 |
Interactive version: theaggregate.ai/benchmark?slug=geocode-leaderboard · How the rankings work · Data refreshed daily, snapshot 2026-07-22.