Gemini native

Native Gemini API

POST/v1beta/models/{model}:generateContent

Compatible with Google's native generateContent protocol, with Bearer auth. Both text and image generation use this endpoint — drop-in for existing Gemini code.

Request body (generateContent)

Field	Type	Required	Description
contents	object[]	Required	Conversation contents, each {role, parts:[{text\|inline_data}]}
systemInstruction	object	Optional	System instruction
generationConfig	object	Optional	Generation params: temperature / maxOutputTokens / topP / topK …
safetySettings	object[]	Optional	Safety filter settings

Text example

bash

curl "https://your-domain/v1beta/models/gemini-2.5-pro:generateContent" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{ "role": "user", "parts": [{ "text": "Hello, Gemini" }] }],
    "generationConfig": { "temperature": 0.7 }
  }'

Image generation

Generate images with the same generateContent endpoint: add responseModalities and imageConfig (aspectRatio, imageSize) to generationConfig. For image-to-image, put the input image as inline_data (mime_type + base64) inside contents[].parts.

Field	Type	Required	Description
generationConfig.responseModalities	string[]	Required	Response modalities. Pass ["IMAGE"] for image generation; use ["TEXT","IMAGE"] to also return text
generationConfig.imageConfig	object	Required	Image generation config (aspectRatio + imageSize)
imageConfig.aspectRatio	string	Required	Aspect ratio: 1:1 / 3:2 / 2:3 / 4:3 / 3:4 / 5:4 / 4:5 / 16:9 / 9:16 / 21:9
imageConfig.imageSize	string	Required	Image quality tier: 1K / 2K / 4K
parts[].inline_data	object	Optional	Image-to-image input: { mime_type, data(base64) } inside contents[].parts

bash

curl "https://your-domain/v1beta/models/gemini-3.1-flash-image-preview:generateContent" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{ "role": "user", "parts": [{ "text": "a neon fox, cinematic" }] }],
    "generationConfig": {
      "responseModalities": ["IMAGE"],
      "imageConfig": { "aspectRatio": "16:9", "imageSize": "2K" }
    }
  }'

Supported models

Image-generation model below; text generation accepts standard Gemini models (e.g. gemini-2.5-pro).

gemini-3.1-flash-image-previewgemini-2.5-flash-image