Skip to main content

OpenAIProvider

openmolt

openmolt / OpenAIProvider

Class: OpenAIProvider

Defined in: providers/OpenAIProvider.ts:23

OpenAI LLM provider. Supports chat-completion models (GPT-4o, GPT-3.5, etc.) and reasoning models (o1, o3 family).

Extends​

Constructors​

Constructor​

new OpenAIProvider(apiKey, baseUrl?): OpenAIProvider

Defined in: providers/OpenAIProvider.ts:30

Parameters​

apiKey​

string

OpenAI API key.

baseUrl?​

string

Override for the API base URL (useful for proxies / Azure).

Returns​

OpenAIProvider

Overrides​

BaseProvider.constructor

Methods​

generate()​

generate(systemPrompt, userMessage, model, config?): Promise<LLMResponse>

Defined in: providers/OpenAIProvider.ts:39

Send a prompt to the underlying LLM and return a normalised response.

Parameters​

systemPrompt​

string

The Maestro system prompt (static across iterations).

userMessage​

string

The per-iteration input-state message.

model​

string

Provider-specific model identifier (e.g. gpt-4o).

config?​

ModelConfig

Optional model-level tuning parameters.

Returns​

Promise<LLMResponse>

Overrides​

BaseProvider.generate