Interface LLMVisionProviderConfig

Configuration for the LLM vision provider.

Example

const config: LLMVisionProviderConfig = {
  provider: 'openai',
  model: 'gpt-4o',
  prompt: 'Describe this image for a search index.',
  apiKey: process.env.OPENAI_API_KEY,
};
interface LLMVisionProviderConfig {
    provider: string;
    model?: string;
    prompt?: string;
    apiKey?: string;
    baseUrl?: string;
}

Properties

provider model? prompt? apiKey? baseUrl?

Properties

provider

provider: string

LLM provider name (e.g. 'openai', 'anthropic', 'google', 'ollama'). Must be resolvable by the generateText() API.

Optional model

model?: string

Model identifier. When omitted, the provider's default vision model is used.

Example

'gpt-4o', 'claude-sonnet-4-20250514', 'gemini-2.0-flash'

Optional prompt

prompt?: string

Custom prompt for image description. When omitted, a default prompt optimized for search indexing is used.

Optional apiKey

apiKey?: string

Override the API key instead of reading from environment variables. Useful for multi-tenant setups where each user has their own key.

Optional baseUrl

baseUrl?: string

Override the provider base URL (e.g. for Ollama or local proxies).

Settings

Member Visibility

Theme

On This Page

providermodelpromptapiKeybaseUrl