Gemini Pro 2.5 Preview demonstrates significant improvements in performance, including top-tier results in math problem-solving, coding, and multi-step reasoning tasks. It’s part of Google's family of scalable Gemini models optimized for AI-native applications and products.
Transformer-based architecture with reasoning-specific layers optimized for multimodal inputs (text and images). Includes advanced post-training enhancements.
Combines billions of tokens with terabytes of multimedia data spanning diverse domains.
The model is available on the AI/ML API platform as "Gemini Pro 2.5 Preview" .
Detailed API Documentation is available here.
Google DeepMind emphasizes ethical considerations in AI development by promoting transparency regarding the model's capabilities and limitations. The organization encourages responsible usage to prevent misuse or harmful applications of generated content.
Gemini Pro 2.5 Preview is available under a commercial license that allows both research and commercial usage rights while ensuring compliance with ethical standards regarding creator rights.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "google/gemini-2.5-pro", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "google/gemini-2.5-pro", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"google/gemini-2.5-pro","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| Intelligence | 16.7 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Coding | 33.3 | Composite score across standardised coding evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Math | 87.7 | Composite score across standardised mathematics evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Humanity's Last Exam | 18.8% | Expert-level questions across many domains | Source | September 22, 2026 |
| GPQA Diamond | 84.0% | Google-proof graduate science questions (hardest subset) | Source | September 22, 2026 |
| AIME 2025 | 86.7% | Competition mathematics (AIME), 2025 | Source | September 22, 2026 |
| LiveCodeBench | 70.4% | Contamination-free competitive programming problems | Source | September 22, 2026 |
| SWE-bench Verified | 63.8% | Resolving verified real GitHub issues | Source | September 22, 2026 |
| MMMU | 81.7% | College-level multimodal understanding + reasoning | Source | September 22, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Gemini Pro 2.5 This page | Coding + agents | |||
| Reasoning + agents | ||||
| Balanced coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Reasoning + agents |
Gemini Pro 2.5 has a 1,000,000 tokens context window and can return up to 65,536 tokens.
Gemini Pro 2.5 takes image, text as input and returns text.
Use google/gemini-2.5-pro as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
Gemini Pro 2.5 became available on September 9, 2025.
Gemini Pro 2.5 is priced at input $1.625 / 1M tokens, output $13 / 1M tokens, cached input $1.625 / 1M tokens.
Yes, Gemini Pro 2.5 can stream responses as they are generated.
Yes, Gemini Pro 2.5 accepts image input alongside text.
Gemini Pro 2.5 was built by Google.
Yes, it supports function calling, including parallel tool calls and structured output.
