← All models

Gemini 2.5 Pro

Published Publisher Googlegemini-2.5-pro

Positioning and Release Date

Gemini 2.5 Pro is Google's multimodal thinking model for complex reasoning, coding, and agentic tasks. Google announced its general release and availability as a stable version on June 17, 2025; the stable version had no model changes from the June 5, 2025 preview. (developers.googleblog.com)

Capabilities and Uses

  • Accepts text, image, audio, video, and PDF inputs and outputs text; suitable for analyzing long documents, datasets, and codebases, as well as addressing mathematics, science, and software development problems.
  • Supports thinking, function calling, code execution, structured output, file search, URL context, context caching, and grounding with Google Search, among other features.
  • The documentation lists an input limit of 1,048,576 tokens and an output limit of 65,536 tokens; the knowledge cutoff is January 2025. (aistudio.google.com)

Limitations and Availability

The model does not support direct image or audio generation or the Live API; generated content may still contain errors, and important results should be independently verified. Google's current documentation marks it as deprecated: the model can no longer be selected in Google AI Studio, and API access is limited to existing users before December 1, 2026. Beginning December 1, 2026, API requests sent to gemini-2.5-pro will be automatically routed to gemini-3.8-flash and billed at the latter's rates; migration also involves API changes, including function calling and thinking parameters. (aistudio.google.com)

Pricing

The following are Google Gemini Developer API paid-tier prices in US dollars (USD) as of the lookup date, November 18, 2026, with all prices expressed per 1 million tokens. Tiers are based on the number of prompt tokens, and output prices include thinking tokens; these are not Vertex AI prices. (aistudio.google.com)

Call method Prompt length Input Output
Standard Up to 200,000 tokens $1.25 $10.00
Standard Over 200,000 tokens $2.50 $15.00
Batch processing Up to 200,000 tokens $0.625 $5.00
Batch processing Over 200,000 tokens $1.25 $7.50

Cached input for standard paid calls costs $0.125 and $0.25 per 1 million tokens for the two prompt tiers above, respectively; cache storage is charged separately at $4.50 per 1 million tokens per hour. The official pricing page also lists a free tier; actual use is subject to quotas and model availability. After automatic routing on December 1, 2026, the receiving model's prices will apply rather than those in the table above. (aistudio.google.com)

Official Sources