Curiso AI is an infinite canvas for your thoughts—a platform that seamlessly connects nodes and AI services to explore ideas in depth without repeating yourself. By guiding the direction of each conversation, Curiso.ai empowers advanced users to unlock richer, more accurate AI interactions. Created by Carsen Klock.
-
Multi OS: Windows, macOS, and Linux app
-
Infinite Canvas: Create and organize your thoughts visually on an unlimited workspace
-
Multiple AI Provider Integration:
- OpenAI
- Anthropic
- xAI
- Groq
- OpenRouter
-
Local AI Inference Provider Integration:
- Ollama
- Exo
- Jan.ai
- LM Studio
- vLLM
-
Custom Model Support: Add and configure custom AI models
-
Model Metrics: View metrics for models like tok/sec and total tokens
-
RAG Support (Retrieval Augmented Generation): Add and configure RAG documents and websites locally
-
Local Transformers.js Embedding Models or OpenAI: Add and configure local embedding models or OpenAI embedding models
-
Local VectorDB IndexedDB: Local VectorDB IndexedDB for RAG
-
Inference Parameters: Customize the inference parameters for your conversations
-
Multiple Boards: Create and manage multiple workspaces
-
Vision Model Support: Add images to your chats for vision models
-
Customizable Interface:
- Theme color selection
- Grid snapping
- Pan and zoom controls
- Double-click zoom functionality
-
Node-Based Conversations: Connect ideas and conversations through an intuitive node chat system
-
Secure: Local encrypted storage of API keys and sensitive data
If you find Curiso.ai useful, please consider donating to support its development.
Bitcoin (BTC) Address: bc1qgamupnnd2v0uj8a5cffyn8d25atahwq3wexue8
Solana (SOL) Address: FLXQhZgyNGgNzE7MEniiHkrh3bs8kfHjd4J1L7KgBWso
Available for Windows, macOS, and Linux. Download latest release
- Bun runtime installed on your system
- API keys for the AI services you plan to use or Ollama installed locally
- Clone the repository:
git clone https://github.com/metaspartan/curiso.git- Navigate to the project directory:
cd curiso- Install dependencies:
bun install- Run the development build:
bun run desktop- On Windows, you will get CORS errors when trying to connect to a local running Ollama instance. Run the command below in command prompt and restart Ollama to resolve this issue.
set OLLAMA_ORIGINS=*Contributions are welcome! Please feel free to submit a pull request. If you have any questions, ideas, or suggestions, please feel free to open an issue.
This project is licensed under the MIT License - see the LICENSE file for details.
Carsen Klock - @metaspartan @carsenklock