Choose theme
AI Resources
Project AIRI
Project AIRI is a companion and virtual-character project built around real-time conversation, voice, animated characters, and an on-desktop presence.
Its public materials cover a browser version, the Stage Tamagotchi desktop app, Live2D and VRM characters, local or hosted model-provider choices, and developing modules for memory, vision, tools, Discord, Minecraft, Factorio, and other interactive work. Use this as a first read, not a recommendation. Open the original project before trusting details like terms, limits, privacy, cost, setup, or safety.
What it is
A character-shaped AI companion
AIRI combines a chat system, speech input and output, model-provider settings, character cards, Live2D or VRM presentation, and optional modules in one companion interface.
Why it stands out
The companion can stay on the desktop
The desktop app can keep an animated character on screen while opening separate chat and settings windows. The wider project also explores voice channels, memory, vision, tools, games, plugins, and computer use rather than stopping at text roleplay.
Availability
Browser, downloads, and self-managed paths
Readers can try the browser version, inspect the public repository and docs, or use release downloads and package-manager routes. The setup can connect to hosted providers or local endpoints such as Ollama, depending on the features and configuration chosen.
Why it matters
What makes it useful
AIRI makes the parts behind an AI companion easier to compare. The character, chat model, speech services, memory, visual presentation, provider connection, and optional game or tool modules are separate choices rather than one hidden bundle. That helps readers see what they would need to run, replace, or keep if they want a companion that feels present beyond a chat tab.
What to know
Where it fits
AIRI sits between a consumer companion app, a virtual-character system, and an experimental agent platform. It is more involved than a simple roleplay chat, but it is not a base model or a promise that every module works locally. The most useful comparison is with other character-driven companions, AI VTuber projects, and interactive desktop agents.
Notable points
What stands out
Desktop character, chat and provider configuration have documented app paths. Several memory, game, computer-use and mobile pieces remain experimental or in progress, so a listed module is not the same as a finished feature on every platform.
Before using
What to review
Which setup path you want: the browser version, a desktop release, a package-manager install, or a self-managed development setup.
Which chat, speech, transcription, and other providers the character will use. AIRI can connect to local endpoints such as Ollama, but other configurations require cloud accounts, API keys, paid services, or internet access.
What leaves your device through the selected providers and modules. A local model choice does not by itself make voice, memory, analytics, accounts, plugins, or connected services fully local or private.
The published AIRI privacy policy describes the mobile Application. It lists analytics and collection such as device and usage details, configured provider names, agent or tool traces, and errors; check its stated scope against the platform and account you use.
Microphone, screen, file, game, Discord, Telegram, MCP, plugin, and computer-use permissions before enabling modules that can observe or act outside the chat window.
Current release notes and module status before depending on the app. AIRI is still pre-1.0, and official materials mark several features as experimental or in progress.
Reader fit
Who may find it relevant
People who want an animated AI companion that can stay present on the desktop instead of living only in a browser tab.
Readers comparing how virtual-character projects combine chat, voice, memory, model providers, character assets, and interactive modules.
Builders studying a large public codebase for AI companions, AI VTubers, Live2D or VRM interfaces, voice pipelines, and game integrations.
Less relevant for someone who wants a guaranteed fully offline, analytics-free companion with no provider setup, account choices, or experimental features.
Editorial note
Why LifeHubber lists it
For a character whose voice matters as much as its conversation, AIRI's VOICEVOX and AivisSpeech paths let a builder choose the speech engine separately while retaining their chosen chat provider. To give a character speech through VOICEVOX or AivisSpeech, use the voice-provider paths added in v0.12.0-beta.5. Configure Speech separately from the chat provider under Consciousness; selecting Ollama for chat does not choose a speech engine. These are beta features, and other enabled connections still have their own data paths.
Source links
Source materials
Reader note
Before relying on this entry
LifeHubber lists entries to help readers inspect AI projects, not to endorse them or prove they are safe, suitable, accurate, maintained, or right for a specific use. We do not verify every entry in depth. Before relying on anything listed, review the original materials, terms, privacy practices, limits, and risks that matter for your situation.
What to explore next
See what sits behind the character.
A companion can feel like one personality while still depending on separate models, tools, memory, and permissions. Understanding those layers makes provider and privacy choices easier to judge.
More in AI Companions
Keep browsing this category
Explore more resources in this category.
SillyTavern
SillyTavern/SillyTavern
A locally installed interface for character-driven AI chats, with personas, lorebooks, group conversations, prompt controls, extensions, chat import and export, and connections to local or hosted model backends.
DLP3D.AI
dlp3d-ai/dlp3d.ai
A real-time framework for voice conversations with customizable 3D characters, combining model providers, speech, memory and reaction logic, facial animation, whole-body motion, and browser playback.
For project maintainers
Listed here? You can use the badge.
If you maintain a project with a current LifeHubber listing, you may add the optional “Listed on LifeHubber AI Resources” badge to its README, docs, or website. No introduction or permission request is needed.
