# Gemini generates image ## OpenAPI Specification ```yaml openapi: 3.0.1 info: title: '' description: '' version: 1.0.0 paths: /v1beta/models/{model}:generateContent: post: summary: Gemini generates image deprecated: false description: >- Gemini Native Format Raw Graph, guide refer to: https://apidoc.cometapi.com/guide-to-calling-gemini-2-5-flash-image-1425263m0 ## 🎉 Nano Banana 2 Image(`gemini-3.1-flash-image-preview`) Now Supported! we highly recommend reading the official Gemini documentation: 👉 https://ai.google.dev/gemini-api/docs/image-generation You only need to replace `https://generativelanguage.googleapis.com` with `https://api.cometapi.com` **Generate images up to 4K resolution Example:** ```bash curl -s -X POST \ "https://api.cometapi.com/v1beta/models/gemini-3-pro-image-preview:generateContent" \ -H "x-goog-api-key: $COMETAPI_KEY" \ -H "Content-Type: application/json" \ -d '{ "contents": [{"parts": [{"text": "Da Vinci style anatomical sketch of a dissected Monarch butterfly. Detailed drawings of the head, wings, and legs on textured parchment with notes in English."}]}], "tools": [{"google_search": {}}], "generationConfig": { "responseModalities": ["TEXT", "IMAGE"], "imageConfig": {"aspectRatio": "1:1", "imageSize": "4K"} } }' | jq -r '.candidates[0].content.parts[] | select(.inlineData) | .inlineData.data' | head -1 | base64 --decode > butterfly.png ``` **Multi-turn image editing Example:** ```bash curl -s -X POST \ "https://api.cometapi.com/v1beta/models/gemini-3-pro-image-preview:generateContent" \ -H "x-goog-api-key: $COMETAPI_KEY" \ -H "Content-Type: application/json" \ -d '{ "contents": [{ "role": "user", "parts": [ {"text": "Create a vibrant infographic that explains photosynthesis as if it were a recipe for a plants favorite food. Show the \"ingredients\" (sunlight, water, CO2) and the \"finished dish\" (sugar/energy). The style should be like a page from a colorful kids cookbook, suitable for a 4th grader."} ] }], "generationConfig": { "responseModalities": ["TEXT", "IMAGE"] } }' > turn1_response.json cat turn1_response.json # Requires jq to parse JSON response jq -r '.candidates[0].content.parts[] | select(.inlineData) | .inlineData.data' turn1_response.json | head -1 | base64 --decode > photosynthesis.png ``` --- ## 🍌 Flash 2.5 Image Updates ### a. Flexible Aspect Ratios Now supports multiple aspect ratio settings for easy content creation across different devices. All resolutions consume 1,290 tokens by default. **Supported aspect ratios:** 1:1, 3:2, 2:3, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9 **Reference examples:** - https://ai.google.dev/gemini-api/docs/image-generation#aspect_ratios - https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-generation#googlegenaisdk_imggen_mmflash_with_txt-drest - https://github.com/GoogleCloudPlatform/generative-ai/blob/main/gemini/getting-started/intro_gemini_2_5_image_gen.ipynb ### b. Model Name Update All new features will be available on the new model ID: **`gemini-2.5-flash-image`**. The previous `gemini-2.5-flash-image-preview` will be deprecated. ⚠️ **Migration required by October 31, 2025** ### c. Force Image Output To address the frequent issue of text-only outputs, you can now set `"responseModalities"` to `["IMAGE"]` only in API requests. This ensures image generation without text-only responses. tags: - 🖼️ Image Models/Gemini parameters: - name: model in: path description: Gemini model ID that supports image generation required: true example: gemini-3.1-flash-image-preview schema: type: string enum: - gemini-3-pro-image - gemini-3.1-flash-image-preview x-apidog-enum: - value: gemini-3-pro-image name: '' description: '' - value: gemini-3.1-flash-image-preview name: '' description: Self-developed model - name: Authorization in: header description: '' required: true example: '{{api-key}}' schema: type: string requestBody: content: application/json: schema: type: object properties: contents: type: array items: type: object properties: parts: type: array items: type: object properties: text: type: string description: >- The specific text content, such as the user's question. x-apidog-orders: - text description: A specific content block, such as a piece of text. description: >- Contains the specific content blocks sent by the role. It is designed as an array to support multimodal input (e.g., including both text and images). x-apidog-orders: - parts description: >- Each item in the array represents a turn (or a message) in the conversation. description: >- Contains the conversation history or the current content to be sent. This is the core field for the conversation. tools: type: array items: type: object properties: google_search: type: object properties: {} x-apidog-orders: [] x-apidog-orders: - google_search generationConfig: type: object properties: responseModalities: type: array items: type: string description: >- (The meaning of this field is uncertain, this is a guess based on the name) Possibly used to specify the expected content modalities (types) for the model's response, for example, to return text only. imageConfig: type: object properties: aspectRatio: type: string description: >- Supports `1:1` `2:3` `3:2` `3:4` `4:3` `4:5` `5:4` `9:16` `16:9` `21:9`, but please refer to the official Gemini documentation for the most accurate information. imageSize: type: string description: Supports `1K`, `2K`, and `4K` parameters. required: - aspectRatio - imageSize x-apidog-orders: - aspectRatio - imageSize required: - responseModalities - imageConfig x-apidog-orders: - responseModalities - imageConfig description: >- Used to configure the model's generation behavior, such as temperature, maximum output length, etc. required: - contents - tools - generationConfig x-apidog-orders: - contents - tools - generationConfig description: The entire request body examples: '1': value: contents: - role: user parts: - text: >- 'Maintain the character features in the image to generate a new portrait photo: a woman leaning on a wooden railing of a traditional Chinese building. She is wearing a blue cheongsam with pink and red floral motifs and a headdress made of colorful flowers, including roses and lilacs. Her right hand gently touches a large kite with a blue background, decorated with pink fish motifs and a pair of large eyes. The background is the interior of an old wooden building, dimly lit and cozy. The painting style is realistic, focusing on the textural details of the clothing patterns, floral headdresses, and wooden buildings generationConfig: responseModalities: - IMAGE imageConfig: aspectRatio: '9:16' summary: text '2': value: contents: - role: user parts: - text: Blend three images to output a high-resolution image - inline_data: mime_type: image/jpeg data: - inline_data: mime_type: image/jpeg data: - inline_data: mime_type: image/jpeg data: generationConfig: responseModalities: - TEXT - IMAGE summary: images to image '3': value: contents: - parts: - text: >- Da Vinci style anatomical sketch of a dissected Monarch butterfly. Detailed drawings of the head, wings, and legs on textured parchment with notes in English. tools: - google_search: {} generationConfig: responseModalities: - TEXT - IMAGE imageConfig: aspectRatio: '1:1' imageSize: 4K summary: 4K '4': value: contents: - parts: - text: >- Visualize the current weather forecast for the next 5 days in San Francisco as a clean, modern weather chart. Add a visual on what I should wear each day tools: - google_search: {} generationConfig: responseModalities: - TEXT - IMAGE imageConfig: aspectRatio: '16:9' imageSize: 1K summary: 2K '5': value: contents: - parts: - text: >- Da Vinci style anatomical sketch of a dissected Monarch butterfly. Detailed drawings of the head, wings, and legs on textured parchment with notes in English. tools: - google_search: {} generationConfig: responseModalities: - TEXT - IMAGE imageConfig: aspectRatio: '1:1' imageSize: 1K summary: 1K responses: '200': description: '' content: application/json: schema: type: object properties: candidates: type: array items: type: object properties: content: type: object properties: role: type: string parts: type: array items: type: object properties: text: type: string x-apidog-orders: - text required: - role - parts x-apidog-orders: - role - parts finishReason: type: string index: type: integer safetyRatings: type: array items: type: string x-apidog-orders: - content - finishReason - index - safetyRatings promptFeedback: type: object properties: safetyRatings: type: array items: type: string required: - safetyRatings x-apidog-orders: - safetyRatings usageMetadata: type: object properties: promptTokenCount: type: integer candidatesTokenCount: type: integer totalTokenCount: type: integer thoughtsTokenCount: type: integer promptTokensDetails: type: 'null' required: - promptTokenCount - candidatesTokenCount - totalTokenCount - thoughtsTokenCount - promptTokensDetails x-apidog-orders: - promptTokenCount - candidatesTokenCount - totalTokenCount - thoughtsTokenCount - promptTokensDetails required: - candidates - promptFeedback - usageMetadata x-apidog-orders: - candidates - promptFeedback - usageMetadata example: candidates: - content: role: model parts: - inlineData: mimeType: image/png data: iVBORw0KGgoAAAA... finishReason: STOP usageMetadata: promptTokenCount: 122 candidatesTokenCount: 1290 totalTokenCount: 1412 trafficType: ON_DEMAND promptTokensDetails: - modality: TEXT tokenCount: 122 candidatesTokensDetails: - modality: IMAGE tokenCount: 1290 modelVersion: gemini-2.5-flash-image-preview createTime: '2025-11-06T07:06:46.306721Z' responseId: BkkMaaHcEsSY4_UP_aOrmAg headers: {} x-apidog-name: Success security: [] x-apidog-folder: 🖼️ Image Models/Gemini x-apidog-status: released x-run-in-apidog: https://app.apidog.com/web/project/810968/apis/api-20873272-run components: schemas: {} securitySchemes: {} servers: - url: https://api.cometapi.com description: Prod Env security: [] ```