> ## Documentation Index > Fetch the complete documentation index at: https://docs.inb.cc/llms.txt > Use this file to discover all available pages before exploring further. # Gemini generate content > Text chat and image / PDF / audio / video recognition using the Gemini native API format ## Models you can use | Model ID | Notes | | ------------------------ | ------------------------------------------------------- | | `gemini-3.1-pro-preview` | Gemini flagship — strong multimodal, very long context | | `gemini-2.5-pro` | High-performance multimodal, up to 1M token context | | `gemini-2.5-flash` | Fast, lightweight — optimized for high-volume workloads | See [`GET /v1/models`](/api-reference/models/list-models) for the full list. ## Multi-turn conversations Alternate `role: user` and `role: model` entries in the `contents` array to build a multi-turn conversation. ## Media recognition Add an `inlineData` entry in `parts` with a `mimeType` (e.g. `image/jpeg`, `application/pdf`, `audio/mpeg`, `video/mp4`) and the base64-encoded file data. ## Streaming Replace `:generateContent` in the path with `:streamGenerateContent?alt=sse` to switch to SSE streaming mode. ## Safety filtering Use `safetySettings` to adjust the blocking threshold for each harm category. ## System instructions Pass a system-level prompt via `systemInstruction`, using the same `parts` format as `contents`. ## OpenAPI ````yaml /openapi/chat.en.yaml post /v1beta/models/{model}:generateContent openapi: 3.1.0 info: title: InfinityBlue API — Chat version: 1.0.0 summary: Unified AI model API gateway — Chat endpoints description: > InfinityBlue is a unified API gateway for AI models. It exposes **OpenAI-, Google Gemini-, and Anthropic Claude-compatible** endpoints, backed by models from OpenAI, Google, DeepSeek, ByteDance (Seedance), Kuaishou (Kling), and more. You don't need to integrate each vendor separately — just point your base URL at InfinityBlue and keep using the official SDK you already know. ## Authentication Every request must include your API key in the header: ``` Authorization: Bearer YOUR_API_KEY ``` Create and manage your API keys in the [console](https://api.getinfinityblue.com/console). ## Endpoint format conventions | Path prefix | Compatible format | | --- | --- | | `/v1/*` | OpenAI (Chat Completions, Responses, Images, etc.) | | `/v1/messages` | Anthropic Claude Messages | | `/v1beta/models/*` | Google Gemini native | ## Model selection Pass any model ID in the `model` parameter. See [`GET /v1/models`](/api-reference/models/list-models) or the [pricing page](https://api.getinfinityblue.com/pricing) for the full list. contact: name: InfinityBlue url: https://getinfinityblue.com servers: - url: https://api.getinfinityblue.com description: Production security: - bearerAuth: [] tags: - name: Chat description: | Text conversations, vision, tool calling, streaming, and reasoning models — all through a single chat endpoint. paths: /v1beta/models/{model}:generateContent: post: tags: - Chat summary: Gemini generate content (text / media recognition) description: > Proxy native Gemini `generateContent` requests, supporting text chat and multimodal recognition (images, PDFs, audio, video). Path format: `/v1beta/models/{model_name}:generateContent` Streaming variant: `/v1beta/models/{model_name}:streamGenerateContent?alt=sse` ⚠️ Media content must be uploaded via `inlineData` as base64-encoded data. `fileData.fileUri` and the Google File API are not supported. operationId: geminiGenerateContent parameters: - name: model in: path description: Model name, e.g. `gemini-2.5-pro` or `gemini-2.5-flash`. required: true schema: type: string examples: - gemini-2.5-pro requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/GeminiGenerateContentRequest' examples: text_chat: summary: Text chat value: contents: - role: user parts: - text: Hi, introduce Gemini in one sentence. multi_turn: summary: Multi-turn conversation value: contents: - role: user parts: - text: Hello - role: model parts: - text: Great to meet you. What would you like to know? - role: user parts: - text: >- I have two dogs in my house. How many paws are in my house? image_recognition: summary: Image recognition value: contents: - role: user parts: - text: Tell me about the instrument in this image. - inlineData: mimeType: image/jpeg data: pdf_recognition: summary: PDF recognition value: contents: - role: user parts: - text: Can you add a few more lines to this poem? - inlineData: mimeType: application/pdf data: audio_recognition: summary: Audio recognition value: contents: - role: user parts: - text: Please describe this audio file. - inlineData: mimeType: audio/mpeg data: video_recognition: summary: Video recognition value: contents: - role: user parts: - text: >- Transcribe the audio from this video and provide visual descriptions. - inlineData: mimeType: video/mp4 data: responses: '200': description: Successful response content: application/json: schema: $ref: '#/components/schemas/GeminiGenerateContentResponse' '400': description: Invalid request parameters content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '401': description: Invalid or missing API key content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '429': description: Rate limit exceeded content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '500': description: Internal server error content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' components: schemas: GeminiGenerateContentRequest: type: object description: >- Gemini generateContent request body, used for both text chat and media recognition. properties: contents: type: array description: | Conversation contents in chronological order. Each element has a `role` (`user` or `model`) and a `parts` array of content fragments. items: type: object properties: role: type: string enum: - user - model description: Message role. parts: type: array description: >- Content parts, which may include text, images, PDFs, audio, or video. items: type: object properties: text: type: string description: Text content. inlineData: type: object description: >- Base64-encoded media data for image, PDF, audio, or video recognition. properties: mimeType: type: string description: | Media MIME type, e.g. `image/jpeg`, `image/png`, `application/pdf`, `audio/mpeg`, `video/mp4`. examples: - image/jpeg data: type: string description: Base64-encoded media file data. generationConfig: type: object description: Generation parameter configuration. properties: temperature: type: number description: Sampling temperature, controlling output randomness. topP: type: number description: Nucleus sampling threshold. topK: type: integer description: >- Top-K sampling — the model considers only the K most probable tokens at each step. maxOutputTokens: type: integer description: Maximum number of tokens to generate in the response. stopSequences: type: array description: Stop sequences — generation halts when any of these is produced. items: type: string safetySettings: type: array description: >- Safety filter settings to adjust blocking thresholds per harm category. items: type: object properties: category: type: string description: Harm category, e.g. `HARM_CATEGORY_HARASSMENT`. threshold: type: string description: Blocking threshold, e.g. `BLOCK_NONE`, `BLOCK_ONLY_HIGH`. tools: type: array description: Tool list for function calling. items: type: object systemInstruction: type: object description: System-level instruction in the same format as a `contents` element. properties: parts: type: array description: System instruction content parts. items: type: object properties: text: type: string description: System instruction text. GeminiGenerateContentResponse: type: object description: Gemini generateContent response body. properties: candidates: type: array description: List of candidate responses generated by the model. items: type: object properties: content: type: object description: Response content. properties: role: type: string description: Role, typically `model`. parts: type: array description: Content parts. items: type: object properties: text: type: string description: Text content. finishReason: type: string description: >- Reason generation stopped, e.g. `STOP`, `MAX_TOKENS`, `SAFETY`. safetyRatings: type: array description: Safety ratings per harm category. items: type: object properties: category: type: string description: Harm category. probability: type: string description: >- Harm probability, e.g. `NEGLIGIBLE`, `LOW`, `MEDIUM`, `HIGH`. usageMetadata: type: object description: Token usage statistics. properties: promptTokenCount: type: integer description: Tokens consumed by the prompt. candidatesTokenCount: type: integer description: Tokens consumed across all candidate responses. totalTokenCount: type: integer description: Total tokens consumed. ErrorResponse: type: object description: Standard error response. properties: error: type: object properties: message: type: string description: Error message. examples: - Invalid duration. Supported range is 4 to 15 seconds. type: type: string description: Error type. examples: - invalid_request_error param: type: - string - 'null' description: The parameter related to the error. examples: - seconds code: type: - string - 'null' description: Error code. examples: - invalid_duration securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: API Key description: > Bearer token authentication, format: `Authorization: Bearer sk-xxxxxx`. Get your API key in the [console](https://api.getinfinityblue.com/console). ````