Gemini 2.5 Pro Goes Live: Paid Tier Now Available for Scaled Production Use!

Exciting news from the Google AI team! Gemini 2.5 Pro, their powerful state-of-the-art model excelling in coding and complex reasoning, has officially launched for scaled, paid usage. This is accessible through the new Gemini 2.5 Pro Preview endpoint. A big congratulations to the Google team on hitting this significant milestone! For developers like us building production-ready applications, especially in demanding fields like AI and crypto, this is a welcome development. Imagine leveraging this scaled performance for real-time analysis of blockchain data, generating complex smart contracts, or powering sophisticated, high-volume automated systems. This launch offers the higher rate limits and performance needed for real-world scale, plus the assurance that usage data on the paid tier won't be used for Google's model improvements. Introducing the "Gemini 2.5 Pro Preview" Paid Tier This tier is specifically designed for applications requiring robust throughput and reliability. Here’s the pricing structure (per 1 million tokens): Modality / Condition Price / 1M tokens Notes Input price ( 200K) $2.50 Text only Output price ( 200K) $15.00 Incl. reasoning tokens (Pricing based on information available April 4, 2025) The paid tier also features significantly increased, tiered rate limits: Tier RPM TPM RPD Tier 1 150 2,000,000 1,000 Tier 2 1,000 5,000,000 50,000 Tier 3 2,000 8,000,000 -- (Rate limits based on information available April 4, 2025. RPD applies specifically to Grounding with Google Search on paid tiers) Key Paid Tier Features: Context Caching: Currently not available. Grounding with Google Search: Includes 1,500 RPD free, then priced at $35 per 1,000 requests. Data Usage: Your prompts and outputs are not used to improve Google's products. Free Tier Access Continues via Experimental Endpoint The free tier for Gemini 2.5 Pro remains available via the gemini-2.5-pro-exp-03-25 endpoint. As confirmed by Google's Logan Kilpatrick, both the paid "Preview" and the free "Experimental" endpoints utilize the exact same underlying model. Key Free Tier Features: Rate Limits: Lower limits apply. Grounding with Google Search: Free of charge, up to 500 RPD. Data Usage: Your prompts and outputs may be used to improve Google's products. Switching Between Models (Example) Using the Google AI SDK for Node.js/Typescript, selecting the model is straightforward: import { GoogleGenerativeAI } from "@google/generative-ai"; // Ensure your API key is set in environment variables or configured securely const genAI = new GoogleGenerativeAI(process.env.API_KEY!); // To use the new paid preview model: const paidModel = genAI.getGenerativeModel({ model: "gemini-2.5-pro-preview", // Add other generationConfig settings as needed }); // To use the free experimental model: const freeModel = genAI.getGenerativeModel({ model: "gemini-2.5-pro-exp-03-25", // Add other generationConfig settings as needed }); async function run() { // Example usage with the paid model const prompt = "Explain the difference between RPM and TPM in API rate limits."; try { const result = await paidModel.generateContent(prompt); const response = result.response; const text = response.text(); console.log(text); } catch (error) { console.error("Error calling the API:", error); } } run(); Why This Matters for Developers The availability of a scalable, paid Gemini 2.5 Pro tier is crucial for building demanding, production-grade AI applications. Having reliable, high-throughput access to a top-tier model like this is a game-changer for complex tasks. What's the first production capability you're planning to build or enhance using the scaled Gemini 2.5 Pro Preview? Let me know! Official Resources: Pricing Details: https://ai.google.dev/gemini-api/docs/pricing Rate Limit Information: https://ai.google.dev/gemini-api/docs/rate-limits Gemini #GoogleAI #AI #LLM #Developer #Tech #MachineLearning #Gemini2.5Pro #API #Cloud #AIServices #CryptoDev #TypeScript #NodeJS

import { GoogleGenerativeAI } from "@google/generative-ai";

// Ensure your API key is set in environment variables or configured securely
const genAI = new GoogleGenerativeAI(process.env.API_KEY!);

// To use the new paid preview model:
const paidModel = genAI.getGenerativeModel({
  model: "gemini-2.5-pro-preview",
  // Add other generationConfig settings as needed
});

// To use the free experimental model:
const freeModel = genAI.getGenerativeModel({
  model: "gemini-2.5-pro-exp-03-25",
  // Add other generationConfig settings as needed
});

async function run() {
  // Example usage with the paid model
  const prompt = "Explain the difference between RPM and TPM in API rate limits.";
  try {
    const result = await paidModel.generateContent(prompt);
    const response = result.response;
    const text = response.text();
    console.log(text);
  } catch (error) {
    console.error("Error calling the API:", error);
  }
}

run();

Related Posts

The Ultimate Guide to Flutter Lists with Bloc : Part 1

Mar 23, 2025  0

Building Brain-Computer Interface Applications

Mar 19, 2025  0

Mar 3, 2025  0