Overview
Geospatial Intelligence IndexNot yet rankedText only · Higher is better
Geospatial Agent IndexNot yet rankedText only · Higher is better
Cost per taskNot yet rankedAgent tasks, at list prices · Lower is better
Time per taskNot yet rankedAt 100 output tokens/s · Lower is better
Technical specification
- Developer
- MiniMax
- Weights
- Open
- Harness
- Terminus-2 2.0.0
- Reasoning setting
- Reasoning
- Context window
- 1,048,576 tokens
- Maximum output tokens
- 65,536
- Image input
- Yes
- Temperature
- 1.0
- Input price per 1M tokens
- $0.30
- Cached input price per 1M tokens
- $0.060
- Output price per 1M tokens
- $1.20
- Map and trace judge
- Claude Sonnet 5.5
Frequently asked questions
Is MiniMax M3 good at spatial reasoning?
We benchmark MiniMax M3 on geospatial agent tasks (GIS and Earth observation work with real data). Its best result is 68 out of 100 on the Geospatial Agent Index, 54th of 92 entries, at Reasoning reasoning in MiniMax Code. See that entry or the leaderboard.
Can MiniMax M3 analyse satellite or map images?
It takes images, but we have not benchmarked it on image tasks yet.
How does MiniMax M3 compare with GLM-5.3?
Its closest rival on the Geospatial Agent Index is GLM-5.3 at 68, so MiniMax M3 (68) is level, within the noise of about 3 points. Compare them side by side.