Article

Unity's AI tools in beta: How to get started

May 5, 2026|5 Min
Artwork for Unity's AI tools in beta. A dark purple diamond on a black background.

Unity’s AI tools in beta include the in-editor AI assistant that’s built for game development and grounded in your project. In this blog series, we’ll introduce key features and workflows available to you in the the AI tool suite, starting with this general overview of what’s included and how to sign up.

Unity’s AI tools are now open beta for developers on Unity 6.0 or later. It is the first in-editor AI suite built specifically for game development workflows – not a general-purpose chatbot bolted onto the side of the editor, but an agent that understands your project's scene graph, GameObjects, components, packages, and target platform in real time.

In this post, we walk through everything that shipped in the open beta: the AI Assistant and its three modes (Ask, Plan, and Agent), the MCP Server, the AI Gateway, and the Generators. We also cover how to install the AI beta and start your free trial.

This content is hosted by a third party provider that does not allow video views without acceptance of Targeting Cookies. Please set your cookie preferences for Targeting Cookies to yes if you wish to view videos from these providers.

What AI tools are offered?

Unity’s AI tools are built into Unity 6 and powered by third-party frontier models. The in-editor AI assistant draws on Unity documentation and best practices, and it reads the live context of your active project – so answers and generated code are more relevant to what you are actually building, not a generic Unity project.

These tools are not Unity Muse. Muse was a separate, deprecated tool that used Unity first-party models. The AI tools in beta are products with a fundamentally different architecture: they run inside the Editor, are designed for agentic workflows, and allow you to work with industry leading models via Unity Cloud.

The open beta ships with three core components:

  • The in-editor AI assistant – an in-editor agent for writing code, building scenes, and answering questions about your project
  • The AI Gateway – a secure bridge for connecting your own preferred AI tools (Claude, GPT, and others) directly into the Editor
  • The MCP Server – Unity's official Model Context Protocol server, which lets AI agents in your IDE read and get full context on your Unity project before making any changes

The in-editor AI Assistant

The in-editor AI Assistant is the primary interface and available as a chat panel inside Unity. It operates in three modes depending on the complexity and autonomy you need.

Ask mode

Ask mode is the starting point for most interactions. Use it to ask questions about Unity, get explanations for console errors, understand how a specific component works, or look up best practices. The in-editor AI assistant draws on Unity's documentation and the context of your active project to give specific, relevant answers.

Example prompts in Ask mode:

  • “Why is my Rigidbody passing through the floor?”
  • “What is the difference between FixedUpdate and Update?”
  • “How do I set up a NavMesh for my current scene?”

Ask mode

Ask mode is the starting point for most interactions. Use it to ask questions about Unity, get explanations for console errors, understand how a specific component works, or look up best practices. The in-editor AI Assistant draws on Unity's documentation and the context of your active project to give specific, relevant answers.

Example prompts in Ask mode:

  • “Why is my Rigidbody passing through the floor?”
  • “What is the difference between FixedUpdate and Update?”
  • “How do I set up a NavMesh for my current scene?”
The in-editor assistant open in Unity. A user asks how to create a physics-based automatic swinging door.
The in-editor assistant panel, showing a sample query.

Plan mode

Plan mode helps you simplify complex tasks by breaking them into smaller steps. Instead of a single answer, the in-editor AI assistant produces a structured plan for a multi-step task – listing what it intends to do before taking any action. This gives you the chance to review, adjust, and approve before anything changes in your project.

Plan mode is useful for tasks like refactoring a system, setting up a new feature from scratch, or reorganizing scene hierarchy – situations where you want visibility into the approach, so you can make an informed decision before execution.

In-Editor screenshot of the in-editor AI Assistant in Plan mode, presenting a workback schedule and reasoning for suggested project changes
Plan Mode providing a multi-step plan for review.

Agent mode

Agent mode gives the in-editor AI assistant the autonomy to execute tasks end to end. It can write C# scripts, modify scene components, create prefabs, and verify that changes behave as intended – all in a single workflow. Every change made by the in-editor AI Assistant is reversible: it allows you to undo steps, and AI-generated assets are tagged with embedded metadata so they are easy to identify and review.

You also control how much autonomy you want your agent to have. Permissions are configurable, so you can restrict the in-editor AI Assistant to read-only context queries, allow it to write scripts but not modify scenes, or give it full autonomy for rapid prototyping.

Unity AI Assistant open in the Unity Editor. Agent Mode follows prompts to makes changes directly to a scene while sharing its reasoning.
Agent Mode making changes in the scene while providing context

MCP Server and AI Gateway

Unity’s AI tools in beta also open up two integration paths for developers who want to connect external tools to the Unity Editor.

Unity MCP server

The Unity MCP (Model Context Protocol) server lets AI agents running in your IDE – such as Claude in VS Code, Cursor, or other MCP-compatible tools – connect directly to your Unity project's runtime context. Without MCP, an external AI assistant only sees your code files. With MCP, it can inspect scene state, read GameObjects and their components, check console logs, and trigger Editor actions. It even gives you project settings context.

Unity's official MCP server is more performant than open-source alternatives, requires no credits to use, and is included in all Unity Pro, Enterprise, and Industry subscriptions, and available to Personal users on a free trial.

Unity MCP Server open in the Unity Editor, listing out the various GPTs that can be integrated using the technology.
The Unity MCP connection flow

Unity AI Gateway

The Unity AI Gateway lets you bring your own model subscriptions into the Editor. If your studio already pays for Claude, GPT-4o, or another frontier model, you can route that directly through the Gateway without consuming any Unity AI credits. The Gateway is designed for teams who want to centralize model access, apply usage controls, and maintain security and auditability for all AI interactions inside Unity.

Zoomed-in view on the MCP Server panel. The user is selecting an agent among the following options: Unity, Codex, Gemini, Cursor, + Add agent
Selecting a GPT through AI Gateway

Unity AI Gateway lets you connect your existing models without using Unity AI credits. It’s included with Unity Pro, Enterprise, and Industry plans. Developers on Unity Personal can access AI Gateway with a free trial.

Generators

Unity’s AI tools in beta introduces a set of asset generators that convert design references and text prompts into high-quality placeholder assets. In the open beta, Generators can produce:

  • 2D sprites, textures, and placeholder art from text descriptions
  • Scene layouts from image references or design mockups
  • Animations or audio from descriptive prompts

All generated assets are tagged with embedded metadata flagging them as AI-generated. Developers are responsible for managing app store declarations and verifying usage rights for generated content. Generators can be disabled entirely in the Unity Dashboard if needed – granular controls are available so you can keep the assistant active while turning off generation.

An AI-generated prefab of a pirate ship in the Unity Editor.
User Generators to quickly create high quality prefabs for prototyping

Getting started with Unity’s AI tools in beta

Setup takes under 10 minutes. Here’s how to set up the beta in your project:

1. Install Unity 6 or newer

Accessing Unity’s AI tools in beta requires Unity 6.0+. Download the latest version from the Unity Hub or the release archive at unity.com/download.

2. Link your project to Unity Cloud

Open the Unity Dashboard and create or link a cloud project. Unity’s AI tools in beta requires cloud connectivity for your project to function.

3. Install the AI packages

Click the AI button in the Editor toolbar. You will be prompted to install the in-editor AI assistant package. If you cannot see the button, follow the installation steps in the Unity AI documentation to install via the Package Manager.

4. Start your free trial

Unity Personal users get 1,000 credits free for 14 days. Pro, Enterprise, and Industry users have Unity AI access included in their existing subscriptions – simply install the package and sign in.

5. Start building

Open the Assistant panel, pick a mode, and type your first prompt. A good starting point: ask it to explain your current scene structure, or generate a simple player controller script.

More on Unity AI

If you’re interested in reading more about what’s included in Unity’s AI tools in beta, we invite you to read other articles in this ongoing series:

Try Unity’s AI tools in beta

The AI tools in beta open beta are available now for all Unity 6.0 developers. Sign up for a free trial, explore the in-editor AI Assistant, connect your preferred tools via the AI Gateway, and start experimenting with what your development workflow looks like with a project-aware AI agent built in.

Sign up and learn more about plans, pricing, and data privacy at unity.com/features/ai

Full documentation is available in the AI documentation linked from the Editor or at docs.unity3d.com.

Unity’s AI tools are currently in open beta. As such, features, behavior, and availability described in this post are under active development and may change, be limited, or be discontinued without notice.