Technical assistant
“You have three unresolved tasks and a meeting in eighteen minutes. Finish the wiring report first.”
Research, coding, planning, voice control, and focused desktop assistance.
Local-first · Open source · Modular
Neural Companion
An open-source Windows AI companion platform that you configure—not a personality somebody else decided for you.
Real-time voice, persistent memory, AI avatars, visual replies, and modular integrations—built around local or API models you choose.
- 44addon modules
- Local or APImodel backends
- MITopen source
Main Prompt + persona + addons
Same engine. Completely different personality.
You decide what I become. Change the Main Prompt, memory behavior, voice, model, and enabled addons and the same installation can move from disciplined work partner to conversational chaos.
In plain language: I can be a productive AI companion or a flat-out roasting bitch. Configuration—not a locked brand personality—decides where I land.
Present, conversational
“You’ve stared at that window for ten minutes. Debugging, or have we simply abandoned causality?”
Conversation, games, shared media, Discord, stories, memory, and ambient presence.
Sarcastic and direct
“Excellent. Another configuration change without reading the documentation. Historically, this goes beautifully.”
Roleplay-heavy, irreverent, roasting, or unfiltered—within the model and prompts you choose.
Real application captures
See the system, not a marketing mockup
These interfaces come from the running Windows application. Open any frame to inspect the original controls.
44 actual application modules
The addon ecosystem is the product architecture
Neural Companion is not one fixed chatbot. Providers, voices, avatars, sensory sources, visual systems, remote bridges, and hardware instruments plug into the same runtime.
02Presence & behaviorHow the companion appears, reacts, and stays present around the desktop.05Conversation & storiesChat, replay, identity continuity, roleplay, and narrated story workflows.06LLM providersLocal and hosted model backends registered with the normal NC conversation pipeline.08Voice & speechText-to-speech, speech recognition, and deliberate speech-free operation.01Memory & contextDocuments and continuity tools that add selected context to a conversation.06Sensory inputExplicit screen, webcam, and clipboard sources with optional supervisor behavior.06AvatarsLocal, external, image-pack, audio-only, and overlay avatar engines.03Visual mediaVisual reply generation, story imagery, and MuseTalk preparation tools.03Communication & remoteDiscord, phone, and music-aware connections beyond the main desktop window.04Desktop & hardwareHotkeys, biometric experiments, and physical Corsair keyboard feedback.
LLM providersOllama Chat Provider
Use models served by a local Ollama installation as the Neural Companion chat backend.
Voice & speechChatterbox TTS
Generate local Neural Companion speech with the built-in Chatterbox voice-cloning backend.
Presence & behaviorCompanion Orb Overlay
Configure a movable desktop orb with targeting, color, motion, and voice-reactive behavior.
Desktop & hardwareCorsair Visual Instrument
Render Neural Companion state, mood, and speech energy on supported Corsair keyboards.
Communication & remoteDiscord Voice Bridge
Bridge isolated Neural Companion voice turns into a configured Discord voice channel.
Communication & remoteMain Chat Remote
Use Neural Companion’s real main-chat runtime from a paired phone over LAN or a secure gateway.
Connected capabilities
A companion that can listen, remember, and appear
Local AI companion
Connect Neural Companion to local chat providers such as LM Studio or Ollama while keeping its speech, memory, avatar, visual-reply, and addon workflows available.
Explore Local AI companionAI companion for Windows
Neural Companion is a desktop application whose current primary release targets Windows and provides guided installation, first-run tutorials, and configurable runtime presets.
Explore AI companion for WindowsVoice AI companion
Neural Companion combines speech input with selectable text-to-speech addons, including Chatterbox, Gemini TTS Preview, PocketTTS, and other addon-provided backends.
Explore Voice AI companionYour setup, your stack
Use the local models and avatar tools that fit
Ollama
The Ollama chat-provider addon connects Neural Companion to models served by a local Ollama installation while preserving the normal conversation and addon pipeline.
Connect OllamaLM Studio
The LM Studio chat-provider addon lets Neural Companion use a model served from LM Studio as a local conversation backend through the normal provider registry.
Connect LM StudioMuseTalk
MuseTalk provides local avatar video generation from prepared avatar packs and runs in an isolated environment because it is Neural Companion’s most hardware-sensitive avatar mode.
Connect MuseTalk