LiteLLM Completion API

Source: https://docs.litellm.ai/docs/completion/input, https://docs.litellm.ai/docs/completion/output, https://docs.litellm.ai/docs/

Purpose

litellm.completion() is the single entry point for chat completions across 100+ LLM providers. It accepts OpenAI Chat Completion params, translates them per-provider, and returns a ModelResponse matching the OpenAI format — regardless of the underlying provider.

When to Use

• Any call to an LLM that should be provider-agnostic (openai/, anthropic/, vertex_ai/, bedrock/, ollama/, azure/, etc.)
• Building apps that swap models without code changes
• Multi-modal requests (text + image + audio + video + documents) across providers
• You need OpenAI-compatible exception types (AuthenticationError, RateLimitError, APIError) from non-OpenAI backends

Basic Usage