Extended Context Window
Processes up to 1,048,576 tokens in a single request, allowing entire codebases, long documents, or extended conversations to be handled without truncation.
Google
Gemini 2.5 Pro Vision
Gemini 2.5 Pro Vision is a multimodal AI model developed by Google DeepMind, designed to reason through complex problems by analyzing text, images, audio, video, and code. It operates as a "thinking model," meaning it works through logical steps before producing a response rather than generating output directly. The model supports a context window of 1,048,576 tokens, enabling it to process large documents, codebases, and extended conversations in a single request. The model is particularly suited for tasks that require combining visual understanding with structured reasoning, such as interpreting diagrams, analyzing image-based data, and generating code from visual inputs. It has demonstrated strong benchmark performance in math, science, and software engineering tasks, including a 63.8% score on the SWE-Bench Verified evaluation. Gemini 2.5 Pro Vision is available through Google AI Studio and via the Gemini API, making it accessible for developers building applications that require both vision and reasoning capabilities.
Jun 17, 2025
1,048,576 context
65,536 tokens output
Extended Context Window
Multimodal Input
Structured Reasoning
Code Generation
Math and Science Tasks
Visual Understanding
Model Overview
High-signal model metadata in a structured two-column overview table.
Provider
The entity that provides this model.
Google
Input Context Window
The number of tokens supported by the input context window.
1,048,576
tokens
Maximum Output Tokens
The number of tokens that can be generated by the model in a single request.
65,536 tokens
tokens
Open Source
Whether the model's code is available for public use.
No
Release Date
When the model was first released.
Jun 17, 2025
11 months ago
Knowledge Cut-off Date
When the model's knowledge was last updated.
June 2025
API Providers
The providers that offer this model. This is not an exhaustive list.
Google, Vertex AI, Gemini API
Modalities
Types of data this model can process.
Text
Image
Video
Audio
Code
What is Gemini 2.5 Pro Vision
A fuller summary of positioning, capabilities, and source-specific details for Gemini 2.5 Pro Vision.
Gemini 2.5 Pro Vision is a multimodal AI model developed by Google DeepMind, designed to reason through complex problems by analyzing text, images, audio, video, and code. It operates as a "thinking model," meaning it works through logical steps before producing a response rather than generating output directly. The model supports a context window of 1,048,576 tokens, enabling it to process large documents, codebases, and extended conversations in a single request.
The model is particularly suited for tasks that require combining visual understanding with structured reasoning, such as interpreting diagrams, analyzing image-based data, and generating code from visual inputs. It has demonstrated strong benchmark performance in math, science, and software engineering tasks, including a 63.8% score on the SWE-Bench Verified evaluation. Gemini 2.5 Pro Vision is available through Google AI Studio and via the Gemini API, making it accessible for developers building applications that require both vision and reasoning capabilities.
Capabilities
What Gemini 2.5 Pro Vision supports
Multimodal Input
Accepts text, images, audio, video, and code as inputs within the same request, enabling cross-modal analysis and generation.
Structured Reasoning
Uses a chain-of-thought approach to work through multi-step problems before producing a final answer, improving accuracy on complex tasks.
Code Generation
Generates and analyzes code across multiple languages, achieving 63.8% on the SWE-Bench Verified benchmark for software engineering tasks.
Math and Science Tasks
Applies logical and quantitative reasoning to solve problems in mathematics and science, with benchmark results reflecting strong performance in these domains.
Visual Understanding
Interprets and reasons over images and video frames, supporting use cases like diagram analysis, chart reading, and image-based question answering.
Pricing for Gemini 2.5 Pro Vision
Primary API pricing shown in the same “quick compare” spirit as the reference page.
Input tokens
$1.25
Per million tokens
Output tokens
N/A
Per million tokens
Price Comparison
Additional usage-cost dimensions synced into the project for this model.
maxTemperature
2
maxResponseSize
65,536 tokens
API Access & Providers
Places where this model is available, based on the synced detail-page metadata.
Google
Vertex AI
Gemini API
Configuration & Parameters
The configurable options currently documented for this model.
Temperature
NumberMax Response Tokens
NumberSupported Request Parameters
Parameters currently listed by OpenRouter or the local catalog for this model.
Model Performance
Benchmark scores synced from the current model source and normalized into the local catalog.
| Benchmark | Score |
|---|---|
|
AIME 2024
American math olympiad problems
|
|
|
GPQA Diamond
PhD-level science questions (biology, physics, chemistry)
|
|
|
HLE
Questions that challenge frontier models across many domains
|
|
|
LiveCodeBench
Real-world coding tasks from recent competitions
|
|
|
MATH-500
Undergraduate and competition-level math problems
|
|
|
MMLU-Pro
Expert knowledge across 14 academic disciplines
|
|
|
SciCode
Scientific research coding and numerical methods
|
Resources & Documentation
Official model cards, release notes, docs, and other references synced from the source page.
Compare Gemini 2.5 Pro Vision with related models
Jump straight into the most relevant side-by-side comparison pages for this model.
Gemini 2.5 Pro Vision vs Gemini 2.5 Pro
Compare Gemini 2.5 Pro Vision and Gemini 2.5 Pro across pricing, context window, capabilities, benchmarks, and API access to choose the better fit for long-context workloads versus long-context workloads.
Gemini 2.5 Pro Vision vs Gemini 2.5 Flash Vision
Compare Gemini 2.5 Pro Vision and Gemini 2.5 Flash Vision across pricing, context window, capabilities, benchmarks, and API access to choose the better fit for long-context workloads versus long-context workloads.
Gemini 2.5 Pro Vision vs Gemini 2.5 Flash Lite
Compare Gemini 2.5 Pro Vision and Gemini 2.5 Flash Lite across pricing, context window, capabilities, benchmarks, and API access to choose the better fit for long-context workloads versus long-context workloads.
Gemini 2.5 Pro Vision vs Gemini 2.5 Flash Image
Compare Gemini 2.5 Pro Vision and Gemini 2.5 Flash Image across pricing, context window, capabilities, benchmarks, and API access to choose the better fit for long-context workloads versus long-context workloads.
Gemini 2.5 Pro Vision vs Gemini 2.5 Flash
Compare Gemini 2.5 Pro Vision and Gemini 2.5 Flash across pricing, context window, capabilities, benchmarks, and API access to choose the better fit for long-context workloads versus long-context workloads.
Gemini 2.5 Pro Vision vs Gemini 1.5 Pro Vision Deprecated
Compare Gemini 2.5 Pro Vision and Gemini 1.5 Pro Vision Deprecated across pricing, context window, capabilities, benchmarks, and API access to choose the better fit for long-context workloads versus general-purpose AI workloads.
FAQ
Common questions about Gemini 2.5 Pro Vision
What is the context window size for Gemini 2.5 Pro Vision?
Gemini 2.5 Pro Vision supports a context window of 1,048,576 tokens, which allows it to process large volumes of text, images, and other inputs in a single request.
What is the training data cutoff for this model?
According to the model metadata, the training date is listed as June 2025.
What input types does Gemini 2.5 Pro Vision support?
The model supports multimodal inputs including text, images, audio, video, and code, making it suitable for tasks that combine visual and language understanding.
Where can I access Gemini 2.5 Pro Vision?
The model is available through Google AI Studio, the Gemini API, and Google Cloud Vertex AI, as well as through MindStudio without requiring separate API key management.
Is Gemini 2.5 Pro Vision suitable for coding tasks?
Yes. The model scored 63.8% on the SWE-Bench Verified evaluation, which measures performance on real-world software engineering tasks, and it supports code generation and analysis across multiple programming languages.
More models from Google
Continue browsing adjacent models from the same provider.