Scores updated · 22 tests both models report · How we compare
Every test, side by side
All 22 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results22 tests
- SATMiMo Embodied 7B by 33.478.745.3+33.4
- RoboBench-PlanningMiMo Embodied 7B by 22.558.736.2+22.5
- RefSpatial-BenchMiMo Embodied 7B by 19.14828.9+19.1
- Where2PlaceMiMo Embodied 7B by 18.663.645+18.6
- RoboSpatial-HomeMiMo Embodied 7B by 16.561.845.3+16.5
- SIBench-miniMiMo Embodied 7B by 11.153.142+11.1
- ShareRobot-Aff.Qwen3 VL 2B by 10.8919.8+10.8
- ShareRobot-Traj.MiMo Embodied 7B by 950.641.6+9
- CV-BenchMiMo Embodied 7B by 8.888.880+8.8
- MMSI-BenchMiMo Embodied 7B by 8.331.923.6+8.3
- MindCubeMiMo Embodied 7B by 7.836.228.4+7.8
- All-Angles-BenchMiMo Embodied 7B by 6.74942.3+6.7
- RoboBench-MCQMiMo Embodied 7B by 6.743.636.9+6.7
- SITE-Bench-VideoMiMo Embodied 7B by 6.758.952.2+6.7
- ERQAMiMo Embodied 7B by 546.841.8+5
- Ego-Plan2MiMo Embodied 7B by 4.439.935.5+4.4
- DA-2KMiMo Embodied 7B by 2.772.269.5+2.7
- SITE-Bench-ImageQwen3 VL 2B by 2.449.952.3+2.4
- 3DSRBenchMiMo Embodied 7B by 2.14239.9+2.1
- ViewSpatialQwen3 VL 2B by 1.136.137.2+1.1
- VSI-Benchtie48.548tie
- EmbSpatial-Benchtie76.275.9tie
Questions people ask
How do you compare the two?
We use the 22 benchmark tests both models have published scores on. None of them is in the eight capability areas we count, so this page lists them without a verdict. Each score is the one shown on the model's own page.