Configuration for Anthropic Claude

Configuration for llama.cpp server

Configuration for Ollama (local LLM server)

LLM provider that connects to an Ollama server via its OpenAI-compatible API. Ollama exposes /v1/chat/completions for chat-style completions.

LLM provider for any server that speaks the OpenAI-compatible `POST /v1/chat/completions` API. OpenAI itself, vLLM, llama.cpp's server and most local OpenAI-compatible runtimes share the exact same wire format — only the base URL, model name and (optional) bearer API key differ — so a single client covers them all instead of one near-identical implementation per vendor.

Configuration for OpenAI-compatible providers

Factory for creating LLM providers

Identifies supported provider platforms

Configuration for vLLM-served models (OpenAI-compatible API)