Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.
Google DeepMind's Gemini 2.5 Pro is a stable multimodal thinking model for complex reasoning and coding tasks, with native support for text, audio, images, video, and PDFs and a 1,048,576-token input context window.
Standard paid tier on ai.google.dev: prompts <=200k tokens are $1.25 input / $10.00 output / $0.125 cached input per 1M tokens. The page also lists higher rates for prompts >200k tokens ($2.50 input / $15.00 output / $0.25 cached input) and cache storage at $4.50 per 1,000,000 tokens per hour.
Google's first-party Gemini API pricing page lists Gemini 2.5 Pro paid-tier standard pricing at $1.25 input, $10.00 output, and $0.125 cached input per 1M tokens for prompts <=200k tokens; the same page shows higher rates for prompts >200k tokens. The model page confirms Gemini 2.5 Pro is multimodal, accepting audio, images, video, text, and PDF inputs.
