# OpenAI GPT-5.4 Models Now Available

OpenAI released GPT-5.4 and GPT-5.4 Pro. We have automatically transitioned all GPT-5.2 users - here is what it means for your analytics.

OpenAI launched **GPT-5.4** and **GPT-5.4 Pro** this week, and we have already updated AI Data Stream to support them. If you were using GPT-5.2 models, you have been **automatically transitioned** - no action needed on your part.

## What Changed for You

**Nothing you need to do.** AI Data Stream automatically migrates your model selection:

- **GPT-5.2** → **GPT-5.4**
- **GPT-5.2 Pro** → **GPT-5.4 Pro**
- **GPT-5.2 Instant** → **GPT-5 Mini**

Your API key, settings, and conversations all remain the same. GPT-5.2 Instant has been retired by OpenAI and replaced by GPT-5 Mini, a faster and more cost-efficient model.

## What Is Better in GPT-5.4

### 1M Token Context Window

GPT-5.4 offers a **1,050,000 token context window** - ideal for feeding large datasets, lengthy conversation histories, or multiple reports into a single request.

### Built-in Computer Use

GPT-5.4 is the first mainline OpenAI model with built-in computer use capabilities. While AI Data Stream does not use this directly, it reflects the model's improved ability to understand and execute multi-step tasks - exactly what happens when the AI gathers data across your connected properties.

### GPT-5.4 Pro

GPT-5.4 Pro uses additional compute to think harder on complex problems, producing smarter and more precise responses. This is ideal for in-depth analytics tasks like correlating data across multiple sources or generating comprehensive reports.

## The Full OpenAI Lineup

With this update, the complete OpenAI model lineup in AI Data Stream is:

- **GPT-5.4** - Most capable, frontier model
- **GPT-5.4 Pro** - Maximum intelligence for complex tasks
- **GPT-5 Mini** - Fast and cost-efficient
- **GPT-4o** - Reliable previous generation
- **GPT-4o Mini** - Budget-friendly option
- **o3** - Deep reasoning
- **o4-mini** - Fast reasoning

## Reasoning Effort Control

GPT-5.4 models support **reasoning effort control**, just like Claude 4.6. This lets you decide how much thinking the model does before answering:

- **Low** - Faster responses, lower cost
- **Medium** - Balanced speed and depth
- **High** - Thorough reasoning for complex analysis

You can set a **team default** in Settings > AI Providers for your OpenAI provider, and override it **per message** using the effort dropdown in the chat interface.

## Switching Models

If you want to try GPT-5.4 or switch between models:

1. Go to **Settings > AI Providers**
2. Click your OpenAI provider
3. Select your preferred model from the dropdown
4. Save

Or simply select it from the model picker when starting a new conversation.

## Questions?

If you have any questions about the transition, [contact us](/contact/) or check our [AI providers documentation](/docs/ai-providers/).