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
- Xiaomi
- 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
- $0.14
- Cached input price per 1M tokens
- $0.0028
- Output price per 1M tokens
- $0.28
- Map and trace judge
- Claude Sonnet 5.5
Frequently asked questions
Is MiMo V2.6 Flash good at spatial reasoning?
We benchmark MiMo V2.6 Flash on geospatial agent tasks (GIS and Earth observation work with real data). Its best result is 71 out of 100 on the Geospatial Agent Index, 41st of 92 entries, at Reasoning reasoning in MiMo Code. See that entry or the leaderboard.
Can MiMo V2.6 Flash analyse satellite or map images?
It takes images, but we have not benchmarked it on image tasks yet.
How does MiMo V2.6 Flash compare with GPT-6 Luna?
Its closest rival on the Geospatial Agent Index is GPT-6 Luna at 71, so MiMo V2.6 Flash (71) is level, within the noise of about 3 points. Compare them side by side.