LlmCore.LLM.Response (llm_core v0.3.0)

Standardized response struct for all LLM providers.

This struct provides a unified format for responses from any provider, whether CLI-based (like Claude Code) or API-based (like OpenAI).

Fields

content - The main text content of the response
provider - Atom identifying the provider (e.g., :claude_code, :openai)
model - String identifying the model used (e.g., "claude-3-opus", "gpt-4")
usage - Map with token usage info (prompt_tokens, completion_tokens, total_tokens)
raw - The raw response from the provider for debugging/passthrough
metadata - Additional provider-specific metadata (latency, request_id, etc.)
structured - Parsed/validated structured output (when requested)
tool_calls - List of tool call requests from the LLM, or nil when no tools were invoked

Example

response = Response.new(
  content: "Hello! How can I help you?",
  provider: :openai,
  model: "gpt-4",
  usage: %{prompt_tokens: 10, completion_tokens: 8, total_tokens: 18}
)

Summary

Types

t()

Functions

new(opts)

Creates a new Response struct from a keyword list.

Types

t()

@type t() :: %LlmCore.LLM.Response{
  content: String.t() | nil,
  metadata: map() | nil,
  model: String.t() | nil,
  provider: atom() | nil,
  raw: map() | nil,
  structured: any() | nil,
  tool_calls: [LlmToolkit.Tool.Call.t()] | nil,
  usage: map() | nil
}

Functions

new(opts)

@spec new(keyword()) :: t()

Creates a new Response struct from a keyword list.

Parameters

opts - Keyword list with response fields

Examples

iex> Response.new(content: "Hello", provider: :openai, model: "gpt-4")
%Response{content: "Hello", provider: :openai, model: "gpt-4", ...}