MiMo-V2.6-Flash Benchmark Details

MiMo-V2.6-Flash currently shows benchmark results led by CyberGym (1 / 11, score 95.10), Terminal-Bench 2.1 (18 / 196, score 87.60), AutomationBench (3 / 20, score 52.30).

Benchmark Results

MiMo-V2.6-Flash

Benchmark Results

AI Agent - Tool Usage

5 evaluations

Benchmark / mode

Score

Rank/total

CyberGym

Thinking EnabledTools

95.10

1 / 11

Terminal-Bench 2.1

Thinking EnabledTools

87.60

18 / 196

OSWorld-Verified

Thinking EnabledTools

80.80

8 / 28

Toolathlon-Verified

Thinking EnabledTools

73.60

9 / 14

Terminal-Bench 4.0

Thinking EnabledTools

28.80

23 / 91

Coding and Software Engineer

1 evaluations

Benchmark / mode

Score

Rank/total

DeepSWE

Thinking EnabledTools

67.90

27 / 89

Agent Level Benchmark

2 evaluations

Benchmark / mode

Score

Rank/total

Job Bench

Thinking EnabledTools

61.20

5 / 8

Agents' Last Exam

Thinking EnabledTools

27.60

11 / 22

Productivity Knowledge

1 evaluations

Benchmark / mode

Score

Rank/total

AutomationBench

Thinking EnabledTools

52.30

3 / 20

Compare with other models