English
chat.gemini_2_5_pro
google/gemini-2-5-pro
chat

The Gemini 2.5 Pro API is Google's flagship thinking model — a 1M-token context window, built-in reasoning, native input for text, images, audio, video, and PDF, function calling, and structured JSON outputs, built for the hardest reasoning, coding, and multimodal tasks. The Gemini 2.5 Pro API is OpenAI-compatible: point any existing SDK at RouterBase and set the model to gemini-2-5-pro. Input at $0.875 / 1M tokens, output at $7.00 / 1M tokens, cache reads at $0.0875 / 1M tokens — 5% below the standard rate, all through one REST endpoint.

Input
Gemini 2.5 Pro
max_tokens4096
temperature1
top_p1
presence_penalty0
frequency_penalty0
Gemini 2.5 Pro Online
Hi! I'm a helpful AI assistant. What can I do for you?

Gemini 2.5 Pro API: Chat

Use the Gemini 2.5 Pro API to run Google's flagship thinking model — top-tier reasoning, coding, and multimodal input with a 1M-token context window.

The Gemini 2.5 Pro API is Google's most capable thinking model — a 1M-token context window, built-in reasoning, native input for text, images, audio, video, and PDF, function calling, and structured JSON outputs, built for the hardest reasoning, coding, and multimodal tasks. Routed through RouterBase, the Gemini 2.5 Pro API is fully OpenAI-compatible: point any existing SDK at RouterBase's base URL, set the model to gemini-2-5-pro, and your first call goes through immediately.

Pricing is $0.875 / 1M input tokens, $7.00 / 1M output tokens, and $0.0875 / 1M cached-input reads — 5% below the standard rate. One RouterBase key — no Google credentials required.

Why this model

Six reasons teams ship on the Gemini 2.5 Pro API

From built-in thinking to 5%-off pricing — what makes the Gemini 2.5 Pro API stand out.

1M-token context

Accepts up to 1 million tokens per request — entire codebases, long document chains, or hours of video transcripts fit in a single Gemini 2.5 Pro API call without chunking.

Built-in thinking

Gemini 2.5 Pro reasons step-by-step before answering, allocating extra compute to difficult sub-problems — higher accuracy on complex coding, math, and analysis tasks.

Native multimodal input

The Gemini 2.5 Pro API understands text, images, audio, video, and PDFs. Pass mixed media in one request to analyze, transcribe, or reason across modalities.

Function calling & JSON

Define tools and request a JSON schema; the Gemini 2.5 Pro API returns structured tool calls and strictly valid JSON for reliable agent and extraction pipelines.

OpenAI-compatible endpoint

The Gemini 2.5 Pro API speaks the OpenAI chat-completions wire format. Point any OpenAI SDK at RouterBase — no Google SDK or separate credentials needed.

One key for 200+ models

The same RouterBase key that calls the Gemini 2.5 Pro API also routes to GPT-5, Claude Opus 4.8, Gemini 3.5 Flash, and 200+ other models — no per-provider credential management.

RouterBase dashboard preview
Quickstart

Get started with the Gemini 2.5 Pro API in 3 steps

From sign-up to your first response in under 5 minutes.

  1. Create a RouterBase API key

    Sign up and generate an API key — one key reaches the Gemini 2.5 Pro API and every other model in the catalog.

  2. Send your first message

    Point any OpenAI-compatible SDK at routerbase.com/v1 and set the model to gemini-2-5-pro. The response follows the standard chat-completions schema — streaming and multimodal input supported.

  3. Inspect usage

    Every response includes a detailed token breakdown — input, output, and cache-read tokens — so cost and cache-hit rate are visible on every call.

Get API Key Read docs
Pricing

Pay only for what you use

RouterBase passes through partner-tier pricing. Compared against the model's official published API rate.

Customer stories

What teams build with the Gemini 2.5 Pro API

Real production loads running on the Gemini 2.5 Pro API and the RouterBase model catalog.

Marcus Reyes
Marcus ReyesCTO, Paradigm AI

The Gemini 2.5 Pro API handles our 800K-token document analysis in one call — built-in thinking, no chunking pipeline. The 5% RouterBase discount is pure margin.

Priya Lakshmi
Priya LakshmiFounder, Quillo

Native multimodal on the Gemini 2.5 Pro API reads video and audio directly — one call transcribes and reasons over the content together.

Thomas Beck
Thomas BeckStaff Engineer, Northbeam

Structured JSON outputs from the Gemini 2.5 Pro API killed our flaky regex parsing. Strictly valid JSON every time, zero retries.

Aoi Tanaka
Aoi TanakaML Lead, Daybreak Robotics

Built-in thinking on the Gemini 2.5 Pro API solved coding tasks our previous model couldn't touch. Accuracy jumped on our internal benchmark.

Jonas Keller
Jonas KellerIndie Developer

1M tokens of context means I can throw the whole repo at it. The Gemini 2.5 Pro API just handles the codebase.

Ethan Nguyen
Ethan NguyenHead of Engineering, Compound Studio

RouterBase puts the Gemini 2.5 Pro API and 200+ other models behind one key. Our team stopped filing requests for new provider accounts.

Sophia Martín
Sophia MartínCTO, Relay

We A/B the Gemini 2.5 Pro API against Gemini Flash per request — same SDK, one model field. The hard cases stay on Pro and everything else gets cheaper.

Lucas Fernandes
Lucas FernandesEngineering Manager, Light

We migrated 42 services to the Gemini 2.5 Pro API over a weekend. The only PR comment was 'wait, that's all?'.

David Okonkwo
David OkonkwoCo-founder, Figment

At $0.875 / 1M input minus 5%, the Gemini 2.5 Pro API gave us flagship reasoning without the flagship bill. The ROI math was instant.

Frequently Asked Questions

Common questions about the Gemini 2.5 Pro API.

It is RouterBase's pass-through to Google's Gemini 2.5 Pro — the flagship thinking model with a 1M-token context, built-in reasoning, native vision/audio/video input, function calling, and structured outputs, served via an OpenAI-compatible REST interface.