# Gemini 3.1 Pro Preview

> A Google reasoning model for software engineering and agentic work, available through RealRelay's OpenAI-compatible endpoint.

## Key facts

- Model author: Google
- Input: $1.9361 per 1M tokens (starting tier)
- Output: $11.6168 per 1M tokens (starting tier)
- Cached input: $0.1936 per 1M tokens
- Context window: 1,048,576 tokens
- Input modalities: audio, file, image, text, video
- Output modalities: text
- Capabilities: reasoning, structured_output, tools, vision
- Released: 2026-02-19
- Model IDs: `jd/gemini-3.1-pro-preview`
- Model page: https://www.realrelay.ai/model/google/gemini-3.1-pro-preview

## Overview

Gemini 3.1 Pro Preview is a Google reasoning model for software engineering, agentic reliability and efficient token usage in complex workflows. It accepts text, image, audio, video and file input and returns text output.

The context window is 1,048,576 tokens, and the model was released on February 19, 2026. Reasoning, structured output, tool calling and vision are supported, and requests use the model ID jd/gemini-3.1-pro-preview on RealRelay.

## Capabilities

Input can be text, images, audio, video or files; every response is text. Reasoning, structured output, tool calling and vision are available.

The 1,048,576-token context window fits large mixed-media prompts and long inputs in a single request. Structured output covers prompts that need a machine-readable reply, and tool calling covers prompts where the model calls functions you define.

## Pricing

Prices are listed per 1M tokens. Input is from $1.94 per 1M tokens and output is from $11.62 per 1M tokens.

Usage is metered: each request is billed on the input and output tokens it uses. Text is the only output modality, so all output usage is text.

## Use Cases

Software engineering assistance, agentic workflows that call tools, and prompts that mix text with images, audio, video or files are natural fits for this model.

Long-context jobs make use of the 1,048,576-token window, and structured output helps when a downstream system expects a formatted reply.

## Call it

OpenAI-compatible chat completions:

```bash
curl https://www.realrelay.ai/v1/chat/completions \
  -H "Authorization: Bearer $REALRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "jd/gemini-3.1-pro-preview", "messages": [{"role": "user", "content": "Hello"}]}'
```

## FAQ

### What is Gemini 3.1 Pro Preview?

It is a Google reasoning model for software engineering and agentic workflows. It supports tool calling, structured output and vision, and accepts text, image, audio, video and file input.

### What are the prices for Gemini 3.1 Pro Preview?

Input is from $1.94 per 1M tokens and output is from $11.62 per 1M tokens.

### How large is the context window?

The context window is 1,048,576 tokens.

### Which inputs and outputs does Gemini 3.1 Pro Preview support?

Text, image, audio, video and file input; output is text only.

### Does Gemini 3.1 Pro Preview support tool calling and structured output?

Yes. Tool calling, structured output and vision are supported alongside reasoning.

### When was Gemini 3.1 Pro Preview released?

It was released on February 19, 2026.

### How do I call Gemini 3.1 Pro Preview through the API?

Use the model ID jd/gemini-3.1-pro-preview in a request to RealRelay's OpenAI-compatible endpoint.