Gemini 2.5 Pro - AI Model API and Pricing

Jobs where the long context earns its rate

Its strengths are a very long context and working well from sources. In practice, you can paste a full interview transcript, research summary or long spec sheet into the prompt. Then ask for copy that stays inside those facts. Shorter models tend to lose details from the middle of a long input. This one is built to hold them.

A blog post that summarizes a whitepaper is a good fit. So is a newsletter built from a month of release notes. The same goes for a product description that must match a technical datasheet line by line.

Open the AI Article Generator and pick the matching preset

It runs in AI Article Generator, which has six presets. They are SEO article, Blog post, Product description, Ad copy, Video script and Newsletter. The tool takes no file uploads, so your source text goes into the prompt itself.

Here is an example for the Blog post preset. "Using only the notes below, write an 800-word post for small clinic owners. Explain the three changes in the new booking rules. Quote the dates exactly as written. Notes: [paste]". For Newsletter, try: "Turn these release notes into a newsletter with one lead story and four short items. Skip anything marked internal."

What a draft costs

Billing rounds up to 1k words. An 800-word blog post costs 16 credits, about $0.17 at the best pack. For 2,000 words the bill is 32 credits, roughly $0.34 at that pack. There is no length setting to adjust, so state the word count you want in the prompt.

Among the 10 text models priced the same way, it ranks 8th from the cheapest. The range runs from 1 credit to 47 credits per 1k words.

When a different model is the better buy

The limit is simple: it costs more than the light models. No source material, just headlines or captions? Claude Haiku 4.5 at 8 credits does that job well. For a long, argued analysis on a tight budget, DeepSeek V4 at 2 credits is far cheaper. Its style can read as technical, though.

Claude Sonnet 5 costs the same 16 credits. Choose it for long editorial pieces where voice and style matter more than staying close to source documents.

Where to actually run it

Gemini 2.5 Pro API

curl -X POST https://ugesi.com/api/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-2-5-pro",
    "prompt": "a product shot on white marble"
  }'

<?php
$ch = curl_init('https://ugesi.com/api/generate');
curl_setopt_array($ch, [
    CURLOPT_POST => true,
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_HTTPHEADER => [
        'Authorization: Bearer YOUR_API_KEY',
        'Content-Type: application/json',
    ],
    CURLOPT_POSTFIELDS => json_encode([
        'model' => 'gemini-2-5-pro',
        'prompt' => 'a product shot on white marble',
    ]),
]);
$response = json_decode(curl_exec($ch), true);
curl_close($ch);
print_r($response);

type GenerateRequest = {
  model: string;
  prompt: string;
};
type GenerateResponse = {
  job_id: number;
  job_token: string;
  credits_held: number;
  status: "queued" | "succeeded";
  output_text?: string;
};

const body: GenerateRequest = {
  model: "gemini-2-5-pro",
  prompt: "a product shot on white marble",
};

const resp = await fetch("https://ugesi.com/api/generate", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.UGESI_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify(body),
});
if (!resp.ok) throw new Error(`HTTP ${resp.status}`);
const job = (await resp.json()) as GenerateResponse;
console.log(job.job_id, job.status);

import requests

resp = requests.post(
    "https://ugesi.com/api/generate",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "model": "gemini-2-5-pro",
        "prompt": "a product shot on white marble",
    },
    timeout=60,
)
resp.raise_for_status()
print(resp.json())

{
  "model": "gemini-2-5-pro",
  "prompt": "a product shot on white marble"
}

The API comes with every credit pack, from Lite to Enterprise, and uses the same balance as the website. API documentation and access →

License and terms of use

Outputs made with paid credits can be used commercially. The maker's own rules for Gemini 2.5 Pro also apply (Google Generative AI Prohibited Use Policy). The full rules are in our Terms of service.

People also look for