GE

Gemini 2.5-Pro

Reasoning modelGemini ProGemini 2.5

Gemini 2.5-Pro

Release date: 2025-06-05Updated: 2025-11-18 23:24:241,960
Live demoGitHubHugging FaceCompare
Parameters
Not disclosed
Context length
1000K
Chinese support
Supported
Reasoning ability

Gemini 2.5-Pro is an AI model published by Google Deep Mind, released on 2025-06-05, for Reasoning model, and 1000K context length, under the 不开源 license, with a 1125.00 score on CodeClash.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

Gemini 2.5-Pro

Model basics

Reasoning traces
Supported
Thinking modes
Thinking modes not supported
Context length
1000K tokens
Max output length
64K tokens
Model type
Reasoning model
Modality (in / out)
Text, Image → Text
Release date
2025-06-05
Model file size
No data
MoE architecture
No
Total params / Active params
No data / N/A
Knowledge cutoff
No data
Gemini 2.5-Pro

Open source & experience

Code license
不开源
Weights license
不开源
GitHub repo
GitHub link unavailable
Hugging Face
Hugging Face link unavailable
Gemini 2.5-Pro

Official resources

Gemini 2.5-Pro

API details

API speed
3/5
No public API pricing yet.
Gemini 2.5-Pro

Benchmark Results

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). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

Thinking
Tool usage

General Knowledge

6 evaluations
Benchmark / mode
Score
Rank/total
86.40
42 / 179
86
20 / 126
58.33
76 / 115
37
47 / 65
21.60
99 / 159
4.90
44 / 59

Common Sense

1 evaluations
Benchmark / mode
Score
Rank/total
54
10 / 45

Coding and Software Engineer

3 evaluations
Benchmark / mode
Score
Rank/total
CodeClash
Standard ModeTools
1125
6 / 8
77.10
34 / 120
67.20
68 / 108

Math and Reasoning

9 evaluations
Benchmark / mode
Score
Rank/total
98.80
1 / 44
92
9 / 62
88
43 / 106
55.20
3 / 16
19
2 / 10
15.20
3 / 9
11
23 / 60
2.10
56 / 80

Writing and Creative Capabilities

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

AI Agent - Tool Usage

2 evaluations
Benchmark / mode
Score
Rank/total
32.60
46 / 46
25.30
28 / 35

Multimodal Understanding

1 evaluations
Benchmark / mode
Score
Rank/total
82
9 / 28

常识推理

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

Agent Level Benchmark

4 evaluations
Benchmark / mode
Score
Rank/total
83.10
4 / 59
Aider-Polyglot
Thinking Mode
79.10
8 / 59

Instruction Following

1 evaluations
Benchmark / mode
Score
Rank/total
49
28 / 29

AI Agent - Information Search

1 evaluations
Benchmark / mode
Score
Rank/total
7.80
44 / 45

Productivity Knowledge

1 evaluations
Benchmark / mode
Score
Rank/total
22
21 / 21

Long Context

1 evaluations
Benchmark / mode
Score
Rank/total
66
8 / 13

Compare with other models

No curated comparisons for this model yet.

Want a custom combination? Open the compare tool

Gemini 2.5-Pro

Publisher

Google Deep Mind
View publisher details
Gemini 2.5-Pro

Model Overview

Gemini 2.5-Pro is an AI model published by Google Deep Mind, released on 2025-06-05, for Reasoning model, and 1000K context length, under the 不开源 license, with a 1125.00 score on CodeClash.

DataLearner on WeChat

Follow DataLearner on WeChat for AI model updates and research notes.

DataLearner WeChat QR code