Nano Banana 2.1 API Guide

export GEMINI_API_KEY="your-api-key"

pip install -U google-genai     # Python SDK
npm install @google/genai       # JavaScript SDK

curl -s -X POST "https://generativelanguage.googleapis.com/v1beta/interactions" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-nano-banana-2.1",
    "input": "Editorial product photo of a matte black coffee bag on a white marble counter, soft window light. The label reads \"ROAST NO.7\" in bold serif type.",
    "response_format": {
      "type": "image",
      "mime_type": "image/png",
      "aspect_ratio": "1:1",
      "image_size": "2K"
    }
  }' > response.json

import base64
from google import genai

client = genai.Client()  # reads GEMINI_API_KEY

interaction = client.interactions.create(
    model="gemini-nano-banana-2.1",
    input=(
        "Editorial product photo of a matte black coffee bag on a white marble counter, "
        'soft window light. The label reads "ROAST NO.7" in bold serif type.'
    ),
    response_format={
        "type": "image",
        "mime_type": "image/png",
        "aspect_ratio": "1:1",
        "image_size": "2K",
    },
)

if not interaction.output_image:
    raise SystemExit(interaction.output_text or "No image returned")

with open("coffee.png", "wb") as f:
    f.write(base64.b64decode(interaction.output_image.data))

import { GoogleGenAI } from "@google/genai";
import { writeFileSync } from "node:fs";

const ai = new GoogleGenAI({}); // reads GEMINI_API_KEY

const interaction = await ai.interactions.create({
  model: "gemini-nano-banana-2.1",
  input:
    "Editorial product photo of a matte black coffee bag on a white marble counter, " +
    'soft window light. The label reads "ROAST NO.7" in bold serif type.',
  response_format: {
    type: "image",
    mime_type: "image/png",
    aspect_ratio: "1:1",
    image_size: "2K",
  },
});

const image = interaction.output_image;
if (!image) throw new Error(interaction.output_text ?? "No image returned");
writeFileSync("coffee.png", Buffer.from(image.data, "base64"));

import base64
from google import genai

client = genai.Client()

with open("portrait.jpg", "rb") as f:
    photo = base64.b64encode(f.read()).decode("utf-8")

interaction = client.interactions.create(
    model="gemini-nano-banana-2.1",
    input=[
        {"type": "text", "text": "Turn this into a 1990s film photo with warm grain. "
                                 "Keep the face, hairstyle and expression unchanged."},
        {"type": "image", "data": photo, "mime_type": "image/jpeg"},
    ],
    response_format={"type": "image", "mime_type": "image/png", "image_size": "2K"},
)

with open("portrait-film.png", "wb") as f:
    f.write(base64.b64decode(interaction.output_image.data))

follow_up = client.interactions.create(
    model="gemini-nano-banana-2.1",
    previous_interaction_id=interaction.id,
    input="Now make the jacket red. Keep everything else exactly the same.",
    response_format={"type": "image", "mime_type": "image/png", "image_size": "2K"},
)

# Upscale the keeper as one more turn. Re-running the original prompt
# at 4K would give a new composition: the model has no seed.
final = client.interactions.create(
    model="gemini-nano-banana-2.1",
    previous_interaction_id=follow_up.id,
    input="Render this exact image again at higher resolution. Change nothing.",
    response_format={"type": "image", "mime_type": "image/png", "image_size": "4K"},
)

import base64
from google import genai

client = genai.Client()

def image_part(path, mime_type="image/png"):
    with open(path, "rb") as f:
        data = base64.b64encode(f.read()).decode("utf-8")
    return {"type": "image", "data": data, "mime_type": mime_type}

interaction = client.interactions.create(
    model="gemini-nano-banana-2.1",
    input=[
        {"type": "text", "text": "The woman from image 1 wearing the jacket from image 2, "
                                 "holding the bottle from image 3, on a rooftop at sunset. "
                                 "Keep her face exactly as in image 1."},
        image_part("woman.png"),
        image_part("jacket.png"),
        image_part("bottle.png"),
    ],
    response_format={"type": "image", "mime_type": "image/png",
                     "aspect_ratio": "4:5", "image_size": "2K"},
)

interaction = client.interactions.create(
    model="gemini-nano-banana-2.1",
    input="A travel poster of the Taipei 101 observatory entrance as it looks today, daylight, wide angle.",
    tools=[{"type": "google_search", "search_types": ["web_search", "image_search"]}],
    response_format={"type": "image", "mime_type": "image/png",
                     "aspect_ratio": "3:4", "image_size": "2K"},
)

interaction = client.interactions.create(
    model="gemini-nano-banana-2.1",
    input="A dense infographic comparing five coffee brewing methods by time, grind size and flavor.",
    generation_config={"thinking_level": "high"},
    response_format={"type": "image", "mime_type": "image/png",
                     "aspect_ratio": "3:4", "image_size": "2K"},
)