Gemini 2.5 Pro
TrendingGoogle · Active · Updated May 18, 2026
Google's most capable thinking model with an industry-leading 1M context window and native multimodal support.
Input Price
$1.25/M
per million tokens
Output Price
$10.00/M
per million tokens
Context Window
1,000,000
tokens
Max Output
8,192
tokens
Technical Specifications
| Provider | |
| Release Date | April 1, 2025 |
| Pricing Type | per token |
| Input Price | $1.25.00 / 1M tokens |
| Output Price | $10.00 / 1M tokens |
| Cached Input | $0.13 / 1M tokens |
| Context Window | 1,000,000 tokens |
| Max Output | 8,192 tokens |
| Input Modalities | text, image, audio, video |
| Output Modalities | text |
| Status | active |
| Availability | api, web_app, mobile_app |
| Latency | medium |
| Rate Limit | 2,000 RPM |
| Pricing URL | View official pricing → |
| Docs URL | — |
Capability Scores
Coding90
Reasoning92
Math91
Image85
Speed75
Overview
Gemini 2.5 Pro is Google DeepMind's frontier thinking model, distinguished by its massive 1 million token context window — the largest of any major model. It can process entire codebases, hours of video, or thousands of pages in a single request. With competitive pricing at $1.25/M input tokens, it offers exceptional value for tasks requiring very long context.
Pros
- +Unmatched 1M token context window
- +Excellent reasoning (92/100) and math (91/100) scores
- +Very competitive pricing for its capability level
- +Native support for video and audio input
Cons
- −Slower inference due to thinking mode overhead
- −Smaller ecosystem and community compared to OpenAI/Anthropic
- −Output capped at 8K tokens
Compare with Alternatives
Use Cases
Processing entire codebases for analysis and refactoring
Long-document research and legal document analysis
Multi-modal reasoning across video, audio, and text