EveryDev.ai
Subscribe
Home
Developers

5,086+ AI companies

  • Radar
  • Trending
AI Tools by Topic
  • AI Coding Assistants
  • Agent Frameworks
  • MCP Servers
  • AI Prompt Tools
  • Vibe Coding Tools
  • AI Design Tools
  • AI Database Tools
  • AI Website Builders
  • AI Testing Tools
  • LLM Evaluations
Follow Us
  • X / Twitter
  • LinkedIn
  • Reddit
  • Discord
  • Threads
  • Bluesky
  • Mastodon
  • YouTube
  • GitHub
  • Instagram
Get Started
  • Users
  • Rate Tools
  • About
  • Editorial Standards
  • Corrections & Disclosures
  • Community Guidelines
  • Advertise
  • Contact Us
  • Newsletter
  • Submit a Tool
  • Start a Discussion
  • Write A Blog
  • Share A Build
  • Terms of Service
  • Privacy Policy
Explore with AI
  • ChatGPT
  • Gemini
  • Claude
  • Grok
  • Perplexity
Agent Experience
  • llms.txt
Theme
With AI, Everyone is a Dev. EveryDev.ai ©
    1. Home
    2. Developers
    3. Juice Boxx Games LLC

    Juice Boxx Games LLC

    Juice Boxx Games is an independent game studio making playful games and software experiences intended to enrich players' lives and bring joy. Its Utsuwa project applies that mission to a privacy-focused, open-source AI companion that users can run locally and customize.

    Visit Website

    At a Glance

    1Tool Listed
    1Product
    12Capabilities
    Discussions
    Columbus, OhioHeadquarters
    2026Est.
    Focus Areas
    Conversational Agents
    3D and Animation
    Local Inference
    Connect
    Latest News
    Utsuwa 0.21.4 released with improved desktop connectivity for local and custom providers, including LM Studio, Together, OpenAI-compatible endpoints, and Windows Ollama.Nov 7, 2026
    Dev Log: thirteen releases brought local OmniVoice, multilingual voice, MCP tools, custom animation and gestures, mood expressions, UI redesign, and reliability hardening.Sep 26, 2026
    Markets
    • Individual AI-companion users
    • Privacy-conscious and local-first AI users
    • VRM/VTuber, anime-character, and virtual-avatar communities
    • Developers and open-source contributors
    • +1 more

    AI Tools by Juice Boxx Games LLC

    (1)
    View Utsuwa
    Utsuwa tool icon

    Utsuwa

    Open Source 3D AI Companion App

    Conversational AI3D and AnimationLocal Inference

    Discussions

    No discussions yet

    Be the first to start a discussion about Juice Boxx Games LLC

    Latest News

    11/07/2026

    Utsuwa 0.21.4 released with improved desktop connectivity for local and custom providers, including LM Studio, Together, OpenAI-compatible endpoints, and Windows Ollama.

    github.com
    09/26/2026

    Dev Log: thirteen releases brought local OmniVoice, multilingual voice, MCP tools, custom animation and gestures, mood expressions, UI redesign, and reliability hardening.

    utsuwa.ai
    07/19/2026

    Utsuwa moved from MIT to AGPL-3.0-or-later beginning with version 0.12.1.

    utsuwa.ai
    07/09/2026

    Dev Log: photo mode, touch reactions, reminders, and a more playful companion experience shipped.

    utsuwa.ai

    Products & Services

    1
    Utsuwa
    2026

    Free, open-source AI companion software available as a browser app and desktop application for macOS, Windows, and Linux. It combines a 3D VRM avatar, chat, voice, lip-sync, semantic memory, relationship progression, expressions, animations, photo mode, reminders, augmented reality, MCP tools, and local-first storage. Users bring their own cloud-model keys or connect local models and speech engines.

    Market Position

    Utsuwa positions itself as an open-source, local-first alternative to closed AI companions such as xAI's Grok Companion/Ani and as a user-owned software alternative to proprietary companion hardware concepts such as Razer's Project Ava. Its differentiators are user-selected avatars and providers, local storage and offline operation, no required account or subscription, AGPL licensing, and support for cloud, local, and OpenAI-compatible models.

    Leadership

    Founders

    CD

    CJ Dyas

    Design engineer, game developer, and IT professional; GitHub identifies him as the maintainer of JuiceBoxxGames/utsuwa and a member of the JuiceBoxxGames organization. His public profile also links RoveIQStudio and describes him as based in Ohio. In the Utsuwa founding post, Dyas writes that he started the project as a personal development experiment.

    Executive Team

    CD

    CJ Dyas

    Founder and lead developer

    Publicly describes himself as a design engineer, game developer, and IT professional. He maintains the Utsuwa repository and writes the project's development and founding posts.

    Founding Story

    CJ Dyas began Utsuwa in early 2026 as a small project to experiment with VRM, LLMs, local-first storage, and an AI companion that users could own rather than rent from a closed platform. The initial vision was an open, privacy-respecting companion whose model, voice, avatar, and data could be chosen and controlled by the user, with everything capable of running locally.

    Business Model

    Revenue Model

    Utsuwa is distributed as free, open-source software under AGPL-3.0-or-later. It has no subscription or paywalled features; users bring their own API keys and pay any third-party model or speech providers directly, or run local models with no provider billing. The hosted web app may relay cloud-provider requests but the company says it does not store messages or keys.

    Pricing Tiers

    Utsuwa
    $0

    Free forever; no subscription, no account, and no paywalled features. Available on the web and as desktop builds for macOS, Windows, and Linux.

    Target Markets

    Industries & Segments
    • Individual AI-companion users
    • Privacy-conscious and local-first AI users
    • VRM/VTuber, anime-character, and virtual-avatar communities
    • Developers and open-source contributors
    • Creators and gamers interested in interactive characters and dating-sim-inspired experiences
    Use Cases
    • Private AI companionship and conversational presence
    • Running an AI companion locally without mandatory accounts, subscriptions, or company-hosted model access
    • Users who want to bring their own LLM, voice provider, avatar, or local model
    • VRM avatar creators, VTuber and character enthusiasts, and interactive-media experimentation
    • Creative roleplay, memory-enabled conversation, visual-novel/dating-sim-inspired interactions, and photo creation
    • Desktop productivity or ambient companionship through a transparent overlay

    Quick Facts

    Headquarters
    Columbus, Ohio, USA
    Founded
    2026
    Entity Type
    Ohio limited liability company
    Office Locations
    Columbus

    History & Milestones

    November 7, 2026

    Released Utsuwa 0.21.4, improving desktop connectivity for local and custom providers, including LM Studio, Together, other OpenAI-compatible endpoints, and Windows Ollama defaults.

    September 26, 2026

    Reported thirteen releases from 0.12.0 through 0.19.2, including local OmniVoice, multilingual voice support, MCP tools, custom animations and gestures, mood expressions, a unified UI, reliability hardening, and a site redesign.

    July 19, 2026

    Moved Utsuwa licensing for version 0.12.1 and later from MIT to AGPL-3.0-or-later to require hosted or modified versions to keep changes open.

    July 9, 2026

    Shipped photo mode, touch reactions, reminders/timers, and related playful companion interactions.

    July 3, 2026

    Reached a fully local-capable stack using local LLMs, local TTS, and local Whisper speech-to-text; added OpenAI-compatible endpoints, photo sharing with vision-capable models, cross-platform desktop builds, and desktop overlay mode.

    Key Capabilities

    12
    3D VRM avatar loading with camera framing, head tracking, facial expressions, blinking, lip-sync, and physics for hair and clothing
    Text chat with streaming responses, 3D speech bubbles, conversation history, and chat-window or immersive modes
    Long-term semantic memory using local embeddings and a memory inspector/graph
    Relationship system with mood, events, time-aware behavior, and eight progression stages
    Voice input through local Whisper servers, Groq, OpenAI Whisper, or browser speech recognition
    Text-to-speech through ElevenLabs, OpenAI TTS, Fish Audio, local OpenAI-compatible engines, and local OmniVoice

    Integrations & Partnerships

    Platform Integrations

    • Web app at app.utsuwa.ai and desktop builds for macOS, Windows, and Linux via Tauri
    • LLM providers: OpenAI, Anthropic, Google Gemini, DeepSeek, xAI Grok, Ollama, LM Studio, and OpenAI-compatible endpoints such as OpenRouter, Together, vLLM, and LiteLLM
    • Speech providers: ElevenLabs, OpenAI TTS, Fish Audio, Kokoro-FastAPI, openedai-speech, OmniVoice, local Whisper servers, Groq, and browser Web Speech API
    • MCP servers including Home Assistant, GitHub, and web-search/server integrations such as SearXNG
    • VRM avatars from sources such as VRoid Hub and Booth; WebXR-capable Android Chrome and headset browsers

    Connect

    Website
    utsuwa.ai
    GitHub
    JuiceBoxxGames

    AI Topics

    3

    Juice Boxx Games LLC focuses on these topics:

    Conversational Agents(1)
    3D and Animation(1)
    Local Inference(1)
    Back to all developersSuggest an edit