Twigg

Twigg is a stateful API that manages all conversation state for you, eliminating the complexity of handling context across different LLM providers. You create a chat once, send only the next event, and Twigg stores the full conversation history, fits it to each model's context window, and routes requests to Anthropic, OpenAI, Google, xAI, Fireworks, or OpenRouter. Every request comes with a cost report so you always know what each turn cost. Twigg acts as a hosted context store, assembler, and model router. Conversations live outside the LLM providers, meaning you can switch models mid-conversation without losing context. The API is provider-agnostic, so you never lock in to a single vendor. With a simple REST API and API key authentication, Twigg integrates directly into any backend or application that needs multi-provider LLM orchestration with minimal effort.