Generate images with Gemini

POST

v1beta

models

{model}

:generateContent

import os
from pathlib import Path
from google import genai
from google.genai import types

client = genai.Client(
    http_options={"api_version": "v1beta", "base_url": "https://api.runbridge.ai"},
    api_key=os.environ["RUNBRIDGE_API_KEY"],
)

response = client.models.generate_content(
    model="gemini-3.1-flash-image",
    contents="A Monarch butterfly anatomical sketch on textured parchment, Da Vinci style",
    config=types.GenerateContentConfig(
        response_modalities=["TEXT", "IMAGE"],
        image_config=types.ImageConfig(aspect_ratio="1:1", image_size="4K"),
    ),
)

final_image = None
for part in response.parts:
    if getattr(part, "thought", False):
        continue
    if part.text:
        print(part.text)
    elif part.inline_data is not None:
        final_image = part.inline_data

if final_image:
    extension = {
        "image/jpeg": "jpg",
        "image/png": "png",
        "image/webp": "webp",
    }.get(final_image.mime_type)
    if not extension:
        raise ValueError(f"Unsupported image MIME type: {final_image.mime_type}")
    filename = f"output.{extension}"
    Path(filename).write_bytes(final_image.data)

{
  "candidates": [
    {
      "content": {
        "role": "<string>",
        "parts": [
          {
            "text": "<string>",
            "inlineData": {
              "mimeType": "image/jpeg",
              "data": "<string>"
            },
            "thought": false,
            "thoughtSignature": "<string>"
          }
        ]
      },
      "finishReason": "STOP",
      "index": 123,
      "safetyRatings": [
        {
          "category": "<string>",
          "probability": "<string>"
        }
      ]
    }
  ],
  "usageMetadata": {
    "promptTokenCount": 123,
    "candidatesTokenCount": 123,
    "totalTokenCount": 123,
    "thoughtsTokenCount": 123
  }
}

For step-by-step examples, see Use Gemini image models.

Use /v1beta/models/{model}:generateContent to generate or edit Nano Banana images. Put the model ID in the URL and send the prompt or source image parts in contents.

Choose a model

Model	Reference example
`gemini-3.1-flash-image`	Nano Banana 2 text-to-image, editing, composition, and 4K image examples.
`gemini-3-pro-image`	Nano Banana Pro with a 2K output example.
`gemini-3.1-flash-lite-image`	Nano Banana 2 Lite; supports 1K output only.

Use these stable IDs in requests. See Google’s Nano Banana guide for model options and the deprecation schedule for model lifecycle dates.

Gemini image responses can include intermediate image parts where thought is true. These are not the final output. When saving generated images, skip thought: true parts and use the last image part where inlineData exists and thought is not true. Choose the saved file extension from its mimeType: .jpg for image/jpeg, .png for image/png, or .webp for image/webp.

Authorizations

Authorization

string

header

required

Use your RunBridge AI API key.

Path Parameters

model

string

default:gemini-3.1-flash-image

required

Nano Banana image model ID. Use gemini-3.1-flash-image for the Flash examples, gemini-3-pro-image for Pro, or gemini-3.1-flash-lite-image for the Lite example.

Body

application/json

contents

object[]

required

Conversation turns. Each item has a role ("user" or "model") and parts array containing text and/or image data.

Show child attributes

generationConfig

object

Controls generation behavior — output modalities, image resolution, thinking, etc.

Show child attributes

tools

object[]

Optional Google Search grounding for gemini-3.1-flash-image and gemini-3-pro-image. Pass [{"google_search": {}}] to use real-time information in generated images. The Lite model does not support search grounding.

Show child attributes

Response

200 - application/json

Success

candidates

object[]

Show child attributes

usageMetadata

object

Show child attributes

Last modified on October 9, 2026

import os
from pathlib import Path
from google import genai
from google.genai import types

client = genai.Client(
    http_options={"api_version": "v1beta", "base_url": "https://api.runbridge.ai"},
    api_key=os.environ["RUNBRIDGE_API_KEY"],
)

response = client.models.generate_content(
    model="gemini-3.1-flash-image",
    contents="A Monarch butterfly anatomical sketch on textured parchment, Da Vinci style",
    config=types.GenerateContentConfig(
        response_modalities=["TEXT", "IMAGE"],
        image_config=types.ImageConfig(aspect_ratio="1:1", image_size="4K"),
    ),
)

final_image = None
for part in response.parts:
    if getattr(part, "thought", False):
        continue
    if part.text:
        print(part.text)
    elif part.inline_data is not None:
        final_image = part.inline_data

if final_image:
    extension = {
        "image/jpeg": "jpg",
        "image/png": "png",
        "image/webp": "webp",
    }.get(final_image.mime_type)
    if not extension:
        raise ValueError(f"Unsupported image MIME type: {final_image.mime_type}")
    filename = f"output.{extension}"
    Path(filename).write_bytes(final_image.data)

{
  "candidates": [
    {
      "content": {
        "role": "<string>",
        "parts": [
          {
            "text": "<string>",
            "inlineData": {
              "mimeType": "image/jpeg",
              "data": "<string>"
            },
            "thought": false,
            "thoughtSignature": "<string>"
          }
        ]
      },
      "finishReason": "STOP",
      "index": 123,
      "safetyRatings": [
        {
          "category": "<string>",
          "probability": "<string>"
        }
      ]
    }
  ],
  "usageMetadata": {
    "promptTokenCount": 123,
    "candidatesTokenCount": 123,
    "totalTokenCount": 123,
    "thoughtsTokenCount": 123
  }
}

​Choose a model

Authorizations

Path Parameters

Body

Response

Choose a model