neograph::llm::OpenAIProvider::Config Struct Reference

Configuration for OpenAI-compatible API connections. More...

#include <openai_provider.h>

Public Attributes
std::string	api_key
	API key for authentication.
std::string	base_url = "https://api.openai.com"
	Base URL of the API endpoint.
std::string	default_model = "gpt-4o-mini"
	Default model name.
int	timeout_seconds = 60
	HTTP request timeout in seconds.

Detailed Description

Configuration for OpenAI-compatible API connections.

Definition at line 41 of file openai_provider.h.

Member Data Documentation

timeout_seconds

int neograph::llm::OpenAIProvider::Config::timeout_seconds = 60

HTTP request timeout in seconds.

Note: NeoGraph's public surface currently uses a mix of int seconds (here, in SchemaProvider, A2AClient default), std::chrono::milliseconds (ACPClient, ACPServer::call_client, async::HttpClient::Options), and dimensionless step counts (RunConfig::max_steps). A future v1.0 standardisation pass will move all timeouts to std::chrono::milliseconds on a process-level NeoGraphConfig defaults struct that individual configs inherit from. Until then, watch the unit when configuring across modules.

Definition at line 55 of file openai_provider.h.

---

The documentation for this struct was generated from the following file:

• include/neograph/llm/openai_provider.h