Token-Oriented Object Notation is a compact, human-readable format for serializing JSON data in LLM prompts. It represents the same objects, arrays, and primitives as JSON, but in a syntax that minimizes tokens and makes structure easy for models to follow.
Stories
TOON (Token-Oriented Object Notation) is a new data interchange format designed to be a faster, smaller, and more efficient alternative to JSON.
Instead of storing field names as text, TOON replaces them with tokens, dramatically reducing payload size and boosting parsing speed across platforms.
It is ideal for real-time applications, high-frequency APIs, multiplayer games, microservices, and systems where performance truly matters.
💡 TOON: Smaller than JSON – Faster than JSON – Built for the future of data exchange.
🕰️ Development History
📌 Concept Stage – Created as a response to the growing overhead of JSON in real-time and high-performance applications.
📌 Early Prototypes – Demonstrated 20–60% reduction in payload size, proving the potential of token-based data mapping.
📌 TOON v1 – Introduced token mapping tables, a defined schema, and cross-language support.
📌 Today – Gaining traction as a lightweight alternative to MessagePack and CBOR while remaining human-readable and easy to implement.
📌 Early Prototypes – Demonstrated 20–60% reduction in payload size, proving the potential of token-based data mapping.
📌 TOON v1 – Introduced token mapping tables, a defined schema, and cross-language support.
📌 Today – Gaining traction as a lightweight alternative to MessagePack and CBOR while remaining human-readable and easy to implement.
⚙️ Key Features
⚡ Ultra-compact format – Tokenized keys dramatically shrink data size.
🚀 Blazing-fast parsing – Token IDs allow near-zero-overhead decoding.
🔒 Safer & more predictable structure – Less brittle than JSON strings.
🌍 Cross-language compatible – Works with JavaScript, Python, Go, Rust, and more.
🔌 Extensible mapping system – Shared token tables keep server and client in sync.
📦 Optimized for streaming – Perfect for WebSockets, live sync, and RPC.
🚀 Blazing-fast parsing – Token IDs allow near-zero-overhead decoding.
🔒 Safer & more predictable structure – Less brittle than JSON strings.
🌍 Cross-language compatible – Works with JavaScript, Python, Go, Rust, and more.
🔌 Extensible mapping system – Shared token tables keep server and client in sync.
📦 Optimized for streaming – Perfect for WebSockets, live sync, and RPC.
🌟 Why Choose TOON?
✅ Smaller payloads = faster responses
✅ High-speed encoding/decoding
✅ Ideal for performance-critical systems
✅ Easy to adopt in existing codebases
✅ Efficient for real-time and bandwidth-limited environments
✅ High-speed encoding/decoding
✅ Ideal for performance-critical systems
✅ Easy to adopt in existing codebases
✅ Efficient for real-time and bandwidth-limited environments
Comment