DataLearner logo

Gemini 2.5-Pro Benchmark Details

Gemini 2.5-Pro currently shows benchmark results led by MATH-500 (1 / 44, score 98.80), Aider-Polyglot (4 / 59, score 83.10), AIME 2024 (9 / 62, score 92). 2 source links are attached for reference.

Benchmark Results

Gemini 2.5-Pro

Benchmark Results

Thinking
Tool usage

General Knowledge

6 evaluations
Benchmark / mode
Score
Rank/total
GPQA Diamond
86.40
45 / 187
MMLU Pro
86
21 / 132
LiveBench
Thinking Level · High
58.33
76 / 115
ARC-AGI
37
49 / 67
HLE
21.60
110 / 170
ARC-AGI-2
4.90
46 / 61

Common Sense

1 evaluations
Benchmark / mode
Score
Rank/total
SimpleQA
54
11 / 47

Coding and Software Engineer

3 evaluations
Benchmark / mode
Score
Rank/total
CodeClash
Standard ModeTools
1125
6 / 8
LiveCodeBench
77.10
34 / 123
SWE-bench Verified
67.20
71 / 111

Math and Reasoning

9 evaluations
Benchmark / mode
Score
Rank/total
MATH-500
98.80
1 / 44
AIME 2024
92
9 / 62
AIME2025
88
44 / 107
IMO-ProofBench
55.20
3 / 16
IMO 2024
19
2 / 10
IMO-ProofBench Advanced
17.60
4 / 8
IMO 2025
15.20
3 / 9
FrontierMath
11
23 / 60
FrontierMath - Tier 4
Standard Mode
2.10
56 / 80

Writing and Creative Capabilities

1 evaluations
Benchmark / mode
Score
Rank/total
Creative Writing
85.85
8 / 23

AI Agent - Tool Usage

2 evaluations
Benchmark / mode
Score
Rank/total
Terminal Bench 2.0
32.60
47 / 47
Terminal-Bench
25.30
28 / 35

Multimodal Understanding

1 evaluations
Benchmark / mode
Score
Rank/total
MMMU
82
10 / 29

Common Sense Reasoning

1 evaluations
Benchmark / mode
Score
Rank/total
Simple Bench
Thinking Enabled
62.40
11 / 63

Agent Level Benchmark

4 evaluations
Benchmark / mode
Score
Rank/total
Aider-Polyglot
32K
83.10
4 / 59
Aider-Polyglot
Thinking Enabled
79.10
8 / 59
τ²-Bench - Telecom
54
32 / 35
Terminal Bench Hard
25
12 / 13

Instruction Following

1 evaluations
Benchmark / mode
Score
Rank/total
IF Bench
49
29 / 30

AI Agent - Information Search

1 evaluations
Benchmark / mode
Score
Rank/total
BrowseComp
7.80
51 / 52

Productivity Knowledge

1 evaluations
Benchmark / mode
Score
Rank/total
GDPval-AA
22
21 / 21

Long Context

1 evaluations
Benchmark / mode
Score
Rank/total
AA-LCR
66
9 / 14
Compare with other models

Sources

kaggle.comkaggle.comartificialanalysis.aiartificialanalysis.ai