GeminiContent

Defined in: adapters/gemini.ts:51

Gemini content structure for messages.

Google Gemini uses a unique parts array format where each message contains an array of content parts. Each part can be text or inline data (images). Role is either ‘user’ or ‘model’ (Gemini calls the assistant ‘model’).

See

• GeminiRequest
• https://ai.google.dev/api/rest/v1/Content

Example

// Text-only message
const textContent: GeminiContent = {
  role: 'user',
  parts: [{ text: 'Hello, Gemini!' }]
};

// Multimodal message with image
const multimodalContent: GeminiContent = {
  role: 'user',
  parts: [
    { text: 'What is in this image?' },
    {
      inlineData: {
        mimeType: 'image/jpeg',
        data: 'base64-encoded-image-data...'
      }
    }
  ]
};

Properties

parts

parts: ({ text: string; } | { inlineData: { data: string; mimeType: string; }; })[]

Defined in: adapters/gemini.ts:56

Array of content parts - can mix text and inline data (images)

---

role

role: "user" | "model"

Defined in: adapters/gemini.ts:53

Message role - ‘user’ or ‘model’ (Gemini’s term for assistant)