LLMAgent.Providers.OpenAI (llm_agent v0.2.0)

View Source

OpenAI provider implementation for LLMAgent.

This module implements the LLM provider interface for OpenAI, handling API calls, response parsing, and error handling specific to OpenAI's API.

Summary

Functions

completion(params)

Sends a completion request to OpenAI's API.

embedding(params)

Generates embeddings for the provided text using OpenAI's API.

Functions

completion(params)

Sends a completion request to OpenAI's API.

Parameters

  • params - A map with parameters for the request:
    • model - The model to use (e.g., "gpt-4")
    • messages - The conversation history
    • tools - Available tools for function calling
    • temperature - Controls randomness (0.0 to 2.0)
    • max_tokens - Maximum tokens to generate

Returns

{:ok, response} on success, {:error, reason} on failure.

Examples 

iex> params = %{
...>   model: "gpt-4",
...>   messages: [%{role: "user", content: "Hello"}],
...>   max_tokens: 500
...> }
iex> {:ok, response} = LLMAgent.Providers.OpenAI.completion(params)
iex> is_map(response)
true

embedding(params)

Generates embeddings for the provided text using OpenAI's API.

Parameters

  • params - A map with parameters for the request:
    • model - The embedding model to use (e.g., "text-embedding-ada-002")
    • input - The text to generate embeddings for

Returns

{:ok, embeddings} on success, {:error, reason} on failure.

Examples 

iex> params = %{
...>   model: "text-embedding-ada-002",
...>   input: "Hello, world!"
...> }
iex> {:ok, embeddings} = LLMAgent.Providers.OpenAI.embedding(params)
iex> is_list(embeddings)
true