DataLearner logo

MiMo-V2.6-Pro Benchmark Details

MiMo-V2.6-Pro currently shows benchmark results led by Terminal-Bench 2.1 (5 / 196, score 89.90), AutomationBench (2 / 20, score 53.10), DeepSWE (14 / 89, score 71.90).

Benchmark Results

MiMo-V2.6-Pro

Benchmark Results

Thinking

AI Agent - Tool Usage

5 evaluations
Benchmark / mode
Score
Rank/total
CyberGym
Thinking EnabledTools
94
2 / 11
Terminal-Bench 2.1
Thinking EnabledTools
89.90
5 / 196
OSWorld-Verified
Thinking EnabledTools
82
6 / 28
Toolathlon-Verified
Thinking EnabledTools
76.90
3 / 14
Terminal-Bench 4.0
Thinking EnabledTools
34.90
18 / 91

Coding and Software Engineer

1 evaluations
Benchmark / mode
Score
Rank/total
DeepSWE
Thinking EnabledTools
71.90
14 / 89

Agent Level Benchmark

2 evaluations
Benchmark / mode
Score
Rank/total
Job Bench
Thinking EnabledTools
62
3 / 8
Agents' Last Exam
Thinking EnabledTools
31.60
7 / 22

Productivity Knowledge

1 evaluations
Benchmark / mode
Score
Rank/total
AutomationBench
Thinking EnabledTools
53.10
2 / 20
Compare with other models