Gemini 2.5 Pro
GeminiBalanced
ThinkingTool UseVisionStructured Output
About this model
Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy...
Performance Tier
Balanced
Gemini 2.5 Pro is a balanced model from Gemini : strong performance at a reasonable price.
Strong cost-performance ratio. Reliable for most professional use cases without premium pricing.
Pricing
This model is included in Elosia plans| Type | per 1M tokens |
|---|---|
| Input (prompt) | $1.25 |
| Output (completion) | $10.00 |
| Image | $1.25 |
| Internal reasoning | $10.00 |
| Cache read | $0.125 |
| Cache write | $0.375 |
Capabilities
Context Length1.0M
Max Output Tokens66K
TokenizerGemini
Inputtext, image, file, audio, video
Outputtext
Release DateJune 17, 2025
Benchmarks
General Intelligence
MMLU
89.8%
GPQA Diamond
84%
Mathematics
MATH-500
90.2%
AIME 2025
83%
Programming
HumanEval
88.5%
SWE-bench Verified
63.8%
Reasoning
ARC-AGI-2
4.9%
Humanity's Last Exam
17.8%
Multimodal
MMMU-Pro
81.7%
Recommended Use Cases
ResearchAnalysisCodingSummarizationData Extraction
Strengths
- Massive 1M token context window
- Strong multimodal capabilities (text, image, video, audio)
- Excellent math and science reasoning (AIME 92%, GPQA 84%)
- Competitive pricing for context size
Limitations
- Can be slower on very long contexts
- Lower software engineering scores than Claude/GPT (SWE-bench 63.8%)
Resources
This model may use your data for training
