# llm_composer v0.19.4 - Table of Contents

## Pages

- [LlmComposer](readme.md)
- [LICENSE](license.md)

- Guides
  - [Providers](providers.md)
  - [Streaming](streaming.md)
  - [Cost Tracking](cost_tracking.md)
  - [Function Calls](function_calls.md)
  - [Provider Router](provider_router.md)
  - [Custom Providers](custom_provider.md)
  - [Configuration Reference](configuration.md)

## Modules

- Core
  - [LlmComposer](LlmComposer.md): `LlmComposer` is responsible for interacting with a language model to perform chat-related operations,
such as running completions and generating responses.
  - [LlmComposer.Function](LlmComposer.Function.md): Defines a struct for representing a callable function within the context of a language model interaction.
  - [LlmComposer.LlmResponse](LlmComposer.LlmResponse.md): Normalized representation of a response coming from any provider.
  - [LlmComposer.Message](LlmComposer.Message.md): Module that represents an arbitrary message for any LLM.
  - [LlmComposer.Provider](LlmComposer.Provider.md): Behaviour for provider modules used by `LlmComposer`.
  - [LlmComposer.Settings](LlmComposer.Settings.md): Defines the settings for configuring chat interactions with a language model.
  - [LlmComposer.StreamChunk](LlmComposer.StreamChunk.md): Normalized representation of a streaming chunk emitted by any provider.

- Providers
  - [LlmComposer.Providers.Bedrock](LlmComposer.Providers.Bedrock.md): Provider implementation for Amazon Bedrock.
  - [LlmComposer.Providers.Bedrock.HttpClient](LlmComposer.Providers.Bedrock.HttpClient.md): ExAws HTTP client for Bedrock using Mint by default, with optional Finch support.
  - [LlmComposer.Providers.Google](LlmComposer.Providers.Google.md): Provider implementation for Google
  - [LlmComposer.Providers.Ollama](LlmComposer.Providers.Ollama.md): Provider implementation for Ollama
  - [LlmComposer.Providers.OpenAI](LlmComposer.Providers.OpenAI.md): Provider implementation for OpenAI chat completions API.
  - [LlmComposer.Providers.OpenAIResponses](LlmComposer.Providers.OpenAIResponses.md): Provider implementation for OpenAI Responses API.
  - [LlmComposer.Providers.OpenRouter](LlmComposer.Providers.OpenRouter.md): Provider implementation for OpenRouter

- Response Parsing
  - [LlmComposer.ProviderResponse](LlmComposer.ProviderResponse.md): Protocol that turns provider-specific raw responses into `LlmComposer.LlmResponse` structs.

- Streaming
  - [LlmComposer.ProviderStreamChunk](LlmComposer.ProviderStreamChunk.md): Protocol that normalizes provider stream payloads into `LlmComposer.StreamChunk` structs.

- Function Calling
  - [LlmComposer.FunctionCall](LlmComposer.FunctionCall.md): Helper struct for function call actions.

  - [LlmComposer.FunctionCallHelpers](LlmComposer.FunctionCallHelpers.md): Helpers for building assistant messages and tool-result messages when handling
function (tool) calls returned by LLM providers.
  - [LlmComposer.FunctionExecutor](LlmComposer.FunctionExecutor.md): Provides manual execution of function calls from LLM responses.

- Cost Tracking
  - [LlmComposer.Cost.CostAssembler](LlmComposer.Cost.CostAssembler.md): Centralized cost information assembly module.
  - [LlmComposer.Cost.Fetchers.ModelsDev](LlmComposer.Cost.Fetchers.ModelsDev.md): models.dev-specific pricing fetcher for OpenAI, Google, and Bedrock providers.
  - [LlmComposer.Cost.Fetchers.OpenRouter](LlmComposer.Cost.Fetchers.OpenRouter.md): OpenRouter-specific pricing fetcher.
  - [LlmComposer.Cost.Pricing](LlmComposer.Cost.Pricing.md): Centralized pricing retrieval and calculation module.
  - [LlmComposer.CostInfo](LlmComposer.CostInfo.md): Struct for tracking costs and token usage of LLM API calls.

- Routing
  - [LlmComposer.ProviderRouter](LlmComposer.ProviderRouter.md): Behaviour for implementing provider routing strategies.
  - [LlmComposer.ProviderRouter.Simple](LlmComposer.ProviderRouter.Simple.md): Simple provider router that implements exponential backoff for failed providers.
  - [LlmComposer.ProvidersRunner](LlmComposer.ProvidersRunner.md): Handles provider execution logic including fallback strategies, routing,
and error handling for multiple provider configurations.

- Cache
  - [LlmComposer.Cache.Behaviour](LlmComposer.Cache.Behaviour.md): Cache behaviour to use other implementations for cache mod.

  - [LlmComposer.Cache.Ets](LlmComposer.Cache.Ets.md): Basic ETS cache.

- Internals
  - [LlmComposer.Errors](LlmComposer.Errors.md): Custom errors in here

  - [LlmComposer.Helpers](LlmComposer.Helpers.md): Provides helper functions for the `LlmComposer` module for handling language model responses.

  - [LlmComposer.HttpClient](LlmComposer.HttpClient.md): Helper mod for setup the Tesla http client and its options

- Exceptions
  - [LlmComposer.Errors.MissingKeyError](LlmComposer.Errors.MissingKeyError.md)