Token Factory API
The Token Factory API gives your applications serverless access to curated open large language models running on dedicated GPUs. There are no servers, GPUs or model weights to manage: you create an API key, choose a model and send requests, and you pay only for the tokens you use.
The API is OpenAI-compatible, so existing clients such as the official OpenAI SDKs for Python and JavaScript, LangChain, LlamaIndex and curl work by changing just the base URL and API key.
Key capabilities include:
Chat completions returned as a single response or streamed token by token over Server-Sent Events.
Tool calling in the standard OpenAI format, so models can request structured function calls from your code.
Reasoning control to switch model thinking on for hard problems or off for fast, low-cost answers.
Safe retries with idempotency keys, so a retried request is never billed twice.
Team API for managing API keys and tracking usage, billing, rate limits and latency programmatically.
Access the Full Developer Guide
Download the complete Token Factory API Developer Guide for comprehensive details on implementing OpenAI-compatible serverless inference. Access the PDF here: Token Factory API Developer Guide.