Skip to main content

Using the API

How you actually talk to an LLM — messages, caching, sampling, streaming, structured output, tools, multimodal.