A 3D companion with a dynamic personality that adapts and responds uniquely to every interaction.

Sign in to add this item to your wishlist, follow it, or mark it as ignored

Early Access Software

Get involved with this software as it develops.

Note: This Early Access software is not complete and may or may not change further. If you are not excited to use this software in its current state, then you should wait to see if it progresses further in development. Learn more

What the developers have to say:

Why Early Access?

“We are launching Aifu in Early Access because, although the core technology is established, we want the community to participate directly in the evolution of its features, interaction tools, and customization options. Early Access allows us to gather direct feedback from users to prioritize which new functions, mechanics, and integrations we should develop next, ensuring that the final product perfectly aligns with what the community is looking for.”

Approximately how long will this software be in Early Access?

“We estimate that Aifu will be in Early Access for approximately 3 to 6 months. This timeframe will depend directly on the feedback we receive from the community and how long it takes us to implement and polish the new features, customization tools, and content expansions that we choose to prioritize together before the official full release.”

How is the full version planned to differ from the Early Access version?

“The full version of Aifu will feature a much richer and more robust ecosystem. We plan to add a wider array of customization options, new interaction mechanics, and secondary tools based on user requests. While Early Access offers the core interactive experience, the final version will feature refined mechanics, expanded content, and systems co-designed alongside our community throughout the process.”

What is the current state of the Early Access version?

“Aifu currently includes its core features, including text and voice interaction, voice synthesis, independent identities with persistent memory, VRM avatar compatibility, image and screenshot analysis, internet searches, timers, and Pomodoro sessions through natural language.

The Early Access version is functional and usable, and covers the features described in the “About This Game” section. However, some areas are still in an early stage, especially the depth of advanced tools, customization options, overall stability, and general polish. We will continue expanding Aifu with new features, interaction improvements, deeper personalization, and adjustments based on community feedback.
”

Will this software be priced differently during and after Early Access?

“Aifu is and will remain a Free-to-Play title. Base access will not change in price after the full release. The application features an in-app purchase system for certain cosmetic or additional features, and we reserve the right to make minor balance adjustments or introduce new options during Early Access to adapt to community feedback and the value of the new content we introduce.”

How are you planning on involving the Community in your development process?

“The community is the central pillar of Aifu's development! We will actively use the Steam Discussion Forums to debate ideas, gather suggestions regarding the app's behavior and customization, and run polls on upcoming features to implement. We want to maintain a completely transparent development process where user feedback and experiences directly dictate the direction of our updates and future content heading toward version 1.0.”
Read more

Use Aifu

Free To Play
 

Report bugs and leave feedback for this software on the discussion boards

See all discussions

About This Software

Aifu is a 3D virtual companion designed to keep you company throughout your digital day. Unlike conventional, flat chatbots, Aifu is built to interact in a much more lifelike way: it can remember your conversations, grow alongside you through persistent memory, and recognize its own visual identity on screen.

Talk Naturally

Type a message or use voice dictation. Hear replies through native Windows speech synthesis or advanced voices, and choose the voice experience that suits you.

Identities and Persistent Memory

Create independent identities with their own personalities and conversation memories. Return to earlier topics across sessions, without sharing one identity’s memories with another. Each identity also knows what its avatar looks like and can talk about its own look and style.

Little Interactions

Give your companion a head pat and watch them react. Small gestures and animated expressions make everyday moments on your desktop a little more playful.

Make Yourself at Home

Click and drag your companion to a new spot on your desktop. Keep them close while you browse, study, or take a break.

Take a Play Break

Challenge Aifu to tic-tac-toe or chess, roll the dice, or leave a decision to a coin toss. Your companion reacts as you play, turning a short break into a shared moment.

Your Companion, Your Look

Import your own avatar, including VRM and unitypackage files, and make it feel at home on your screen. Depending on the model, you can customize different parts of its look. With any avatar, you can also adjust settings like the camera and lighting.

Free Daily Use

Enjoy up to 30 free messages every day to chat, connect, and spend time with Aifu.

A Little Help When You Need It

Share an image or screenshot for Aifu to discuss. Ask for an internet search, set a timer, or start a Pomodoro session in natural language. Optional desktop tools and integrations are used only when you enable them.

Agentic Capabilities and Vision

Aifu doesn’t just talk—it executes tasks to support your workflow:

  • Active Assistance: Request real-time internet searches to resolve queries or manage timers and Pomodoro sessions using natural language.

  • Vision Capabilities: Aifu can analyze what you see. Provide screenshots or paste images directly into the chat for your companion to process, analyze, and comment on.

By Fans, for Fans

We’re AifuStudio, a small independent team who love anime and technology. Aifu began as a companion we wanted on our own desktops. During Early Access, we’re improving the experience with feedback from the community.

Our Cost Philosophy

We want to be 100% transparent: our main goal is not to make a profit. In order for Aifu’s responses to be smart, fluid, and backed by real memory, we rely on infrastructure providers and external high-performance AI servers that charge us for every single interaction.

The application’s pricing model is strictly designed to make the project sustainable and viable over time, covering these direct technical costs. Every single coin collected is reinvested into keeping the servers running and funding the development of new features. Thank you for helping us bring this dream to life!

Avatar Credits: The avatar shown is Mori, created by Torinyan, with concept design by Limebreaker. Mori is a third-party avatar used to demonstrate Aifu.

AI Generated Content Disclosure

The developers describe how their game uses AI Generated Content like this:

Aifu uses real-time generative artificial intelligence to create dynamic text dialogues and voice responses for interactive 3D companions. Content is generated instantly based on the player's interactions and messages, enabling a unique and personalized conversational experience. Language models are technically constrained to maintain the virtual assistant's entertainment focus.

Aifu also includes optional agentic capabilities that, when explicitly enabled by the player, can use approved local tools and user-configured integrations to assist with selected desktop tasks.

Mature Content Description

The developers describe the content like this:

Aifu features interactive chat experiences with AI-driven 3D companions. The conversations may occasionally include mature humor, suggestive or flirtatious themes, and adult-oriented jokes depending on the interaction. Additionally, characters may wear revealing outfits, form-fitting clothing, or swimwear. There is no explicit graphic nudity or sexual violence in the application.

System Requirements

Windows
macOS
SteamOS + Linux
    Minimum:
    • OS: Windows 10 (64-bit)
    • Processor: Dual-core x64: Intel Core i3-4170, AMD Ryzen 3 1200 or better.
    • Memory: 4 GB RAM
    • Graphics: WebGL and DirectX 11 compatible GPU: Intel HD Graphics 4000, AMD Radeon HD 7750 or better.
    • DirectX: Version 11
    • Network: Broadband Internet connection
    • Storage: 1500 MB available space
    • Additional Notes: Requires WebGL hardware acceleration. Microphone and camera are optional. Internet is required for login, chat and online voice services.
    Recommended:
    • OS: Windows 11 64-bit
    • Processor: Quad-core x64: Intel Core i5-8400, AMD Ryzen 5 3600 or better.
    • Memory: 8 GB RAM
    • Graphics: Intel Iris Xe, NVIDIA GeForce GTX 1050, AMD Radeon RX 6400 or better
    • DirectX: Version 11
    • Network: Broadband Internet connection
    • Storage: 3 GB available space
    • Additional Notes: Requires WebGL hardware acceleration. Microphone and camera are optional. Internet is required for login, chat and online voice services.
    Minimum:
    • Requires an Apple processor
    • OS: macOS 12 Monterey or later
    • Processor: Intel Core i5/i7 or Apple Silicon
    • Memory: 4 GB RAM
    • Graphics: Metal / WebGL compatible GPU
    • Network: Broadband Internet connection
    • Storage: 1500 MB available space
    • Additional Notes: Requires WebGL hardware acceleration. Microphone and camera are optional. Internet is required for login, chat and online voice services.
    Recommended:
    • Requires an Apple processor
    • OS: macOS 13 Ventura or later
    • Processor: Intel Core i5/i7 or Apple Silicon
    • Memory: 8 GB RAM
    • Graphics: Metal compatible GPU
    • Network: Broadband Internet connection
    • Storage: 3 GB available space
    • Additional Notes: Requires WebGL hardware acceleration. Microphone and camera are optional. Internet is required for login, chat and online voice services.
    Minimum:
    • OS: Ubuntu 18.04+, Debian 10+ or Fedora 32+
    • Processor: Dual-core x64
    • Memory: 4 GB RAM
    • Graphics: WebGL / OpenGL GPU with active drivers
    • Network: Broadband Internet connection
    • Storage: 1500 MB available space
    • Additional Notes: Requires WebGL hardware acceleration. Microphone and camera are optional. Internet is required for login, chat and online voice services.
    Recommended:
    • OS: Ubuntu 22.04+, Debian 12+, Fedora 38+ or equivalent
    • Processor: Quad-core x64
    • Memory: 8 GB RAM
    • Graphics: Mesa/NVIDIA/AMD with up-to-date drivers
    • Network: Broadband Internet connection
    • Storage: 3 GB available space
    • Additional Notes: Requires WebGL hardware acceleration. Microphone and camera are optional. Internet is required for login, chat and online voice services.
Review Filters