---
title: "Gemini GenerateContent API"
description: "Send standard or streaming Gemini-format requests through Infistar."
lastUpdated: 2026-10-22
---

Use a model whose current catalog entry supports the Gemini format. Authenticate with `Authorization: Bearer YOUR_API_KEY`.

| Request | International endpoint |
| --- | --- |
| Standard | `POST https://infistar.ai/v1beta/models/{model}:generateContent` |
| Streaming | `POST https://infistar.ai/v1beta/models/{model}:streamGenerateContent?alt=sse` |

Replace `{model}` with an exact ID from your key's `GET /v1/models` response. The ID belongs in the URL path.

| Body field | Meaning |
| --- | --- |
| `contents` | Required conversation content, with text or multimodal `parts` |
| `generationConfig` | Optional generation settings such as temperature, topP and maxOutputTokens |
| `systemInstruction` | Optional system instruction |
| `safetySettings` | Optional safety settings supported by the model |

Optional fields and capabilities depend on the selected model. For address and authentication problems, see [Gateway configuration](/en/integration-guides/gateway-config).