> ## 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. # NanoBanana native format > Call NanoBanana via the Gemini native protocol with multi-reference and high-resolution support ## Models you can use | Model ID | Notes | | ---------------- | --------------------------------------------------------------------------------------------- | | `nanobanana` | Standard — balanced speed and quality | | `nanobanana_pro` | High-quality — ideal for product images and precise style unification | | `nanobanana_2` | 2nd generation — budget tier **only supports the OpenAI chat format** (see separate endpoint) | ## Multiple reference images Append multiple `inline_data` objects consecutively inside `contents[].parts`. Each object holds the Base64 data of one reference image, enabling multi-reference input. ## Resolution configuration Inside `generationConfig.imageConfig`: * `aspectRatio`: `1:1`, `3:4`, `4:3`, `9:16`, `16:9` * `imageSize`: `1K`, `2K`, `4K` ## Thinking content Set `generationConfig.thinkingConfig.includeThoughts: true` to include the model's reasoning trace in the response. ## OpenAPI ````yaml /openapi/images.en.yaml post /v1beta/models/{model}:generateContent openapi: 3.1.0 info: title: InfinityBlue API — Images version: 1.0.0 summary: Image generation and editing endpoints description: > InfinityBlue image endpoints support both the OpenAI native format and the NanoBanana (Gemini native) format, covering text-to-image generation, image editing, and multi-reference image fusion. ## 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/images/*` | OpenAI Images API | | `/v1/chat/completions` | OpenAI Chat Completions (NanoBanana OpenAI format) | | `/v1beta/models/{model}:generateContent` | Google Gemini native (NanoBanana native format) | contact: name: InfinityBlue url: https://getinfinityblue.com servers: - url: https://api.getinfinityblue.com description: Production security: - bearerAuth: [] tags: - name: Images description: | OpenAI-compatible image generation and editing endpoints supporting models such as `gpt-image-2`. - name: NanoBanana description: | NanoBanana image generation endpoints available in both Gemini native format and OpenAI chat format, supporting text-to-image, multi-reference fusion, and style-unification workflows. paths: /v1beta/models/{model}:generateContent: post: tags: - NanoBanana summary: NanoBanana image generation (native format) description: | Generate images using NanoBanana models via the Gemini native protocol. Supports text-to-image, single-reference editing, and multi-reference fusion. Configure the output aspect ratio and resolution level in `generationConfig.imageConfig`. Pass the NanoBanana model ID as the `model` path parameter, e.g. `nanobanana_pro`. operationId: createNanoBananaImageNative parameters: - name: model in: path required: true description: | NanoBanana model ID, e.g. `nanobanana_pro`. Accepted values: `nanobanana`, `nanobanana_pro`. schema: type: string example: nanobanana_pro requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/NanoBananaRequest' examples: text_to_image: summary: Text-to-image (basic) value: contents: - role: user parts: - text: Draw a cat sitting on a rooftop under moonlight generationConfig: responseModalities: - TEXT - IMAGE imageConfig: aspectRatio: '16:9' imageSize: 4K with_thinking: summary: Include thinking content value: contents: - role: user parts: - text: Draw a cat sitting on a rooftop under moonlight generationConfig: thinkingConfig: includeThoughts: true responseModalities: - TEXT - IMAGE imageConfig: aspectRatio: '16:9' imageSize: 4K multi_reference: summary: Multi-reference image fusion value: contents: - role: user parts: - text: >- Combine these two reference images to generate a unified-style product promotional image, preserving the main subject's silhouette, with a more premium, clean, and bright overall look - inline_data: mimeType: image/jpeg data: - inline_data: mimeType: image/jpeg data: generationConfig: responseModalities: - TEXT - IMAGE imageConfig: aspectRatio: '1:1' imageSize: 2K responses: '200': description: Image generated successfully content: application/json: schema: $ref: '#/components/schemas/NanoBananaResponse' '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: NanoBananaRequest: type: object required: - contents - generationConfig description: NanoBanana native-format image generation/editing request body. properties: contents: type: array description: | A list of conversation turns. Typically one `user` message; its `parts` array can contain a text prompt and one or more reference images. items: $ref: '#/components/schemas/NanoBananaContent' generationConfig: $ref: '#/components/schemas/NanoBananaGenerationConfig' NanoBananaResponse: type: object description: NanoBanana native-format response body. properties: candidates: type: array description: List of generated candidate results. items: type: object properties: content: type: object description: Content of the candidate result. properties: role: type: string description: Role of the message author (typically `model`). parts: type: array description: Content parts containing text and/or image data. items: type: object description: >- A single content part containing text or inline image data. finishReason: type: string description: The reason generation stopped. safetyRatings: type: array description: Safety rating list for the candidate. items: type: object usageMetadata: type: object description: Token usage statistics. properties: promptTokenCount: type: integer description: Tokens consumed by the prompt. candidatesTokenCount: type: integer description: Tokens consumed by the candidates. 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 size. Supported sizes are 1024x1024, 1536x1024, 1024x1536. type: type: string description: Error type. examples: - invalid_request_error param: type: - string - 'null' description: The parameter related to the error. examples: - size code: type: - string - 'null' description: Error code. examples: - invalid_size NanoBananaContent: type: object required: - role - parts description: A single conversation turn. properties: role: type: string description: >- The role of the message author. Typically `user` for image generation requests. examples: - user parts: type: array description: | A list of content parts, supporting mixed text and image data. The first part is usually a text prompt (`text`); subsequent parts can be `inline_data` objects representing reference images. items: $ref: '#/components/schemas/NanoBananaPart' NanoBananaGenerationConfig: type: object required: - responseModalities - imageConfig description: >- Generation configuration controlling output modalities, image dimensions, and optional thinking content. properties: thinkingConfig: type: object description: >- Configuration for returning model thinking content. Optional for image generation. properties: includeThoughts: type: boolean description: >- Whether to return the model's reasoning trace. Set to `true` to include it. examples: - false responseModalities: type: array description: | List of response modalities specifying what data types to return. Common values for image generation: `TEXT` (caption/description) and `IMAGE` (the generated image). items: type: string enum: - TEXT - IMAGE examples: - - TEXT - IMAGE imageConfig: $ref: '#/components/schemas/NanoBananaImageConfig' NanoBananaPart: type: object description: | A single content part. Use either `text` for a text prompt or `inline_data` for image data — not both in the same part object. To supply multiple reference images, add multiple `inline_data` parts consecutively. properties: text: type: string description: | Text prompt describing the generation target, style, composition, or subject-preservation requirements. examples: - >- Combine these two reference images into a unified-style product promotional image, preserving the main subject silhouette, with a more premium and clean look inline_data: $ref: '#/components/schemas/NanoBananaInlineData' NanoBananaImageConfig: type: object required: - aspectRatio - imageSize description: Image output configuration specifying aspect ratio and resolution level. properties: aspectRatio: type: string description: | Output image aspect ratio. Choose based on the intended use case. Options: `1:1`, `3:4`, `4:3`, `9:16`, `16:9`. examples: - '1:1' - '16:9' imageSize: type: string description: | Output image resolution level. Higher values produce sharper images at the cost of longer generation time. Options: `1K`, `2K`, `4K`. examples: - 2K - 4K NanoBananaInlineData: type: object required: - mimeType - data description: | Inline image data for passing reference or input images. Supply multiple `inline_data` parts within the same `parts` array for multi-reference input; the model synthesizes all images with the text prompt. properties: mimeType: type: string description: >- Image MIME type, e.g. `image/jpeg`, `image/png`, `image/webp`. Must match the actual format. examples: - image/jpeg data: type: string description: | Base64-encoded image data, without the `data:image/...;base64,` prefix. For multi-image input, each image corresponds to a separate `inline_data.data` field. examples: - 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). ````