Image generation and multimodal prompts examples with Google AI Studio

image_generation_with_gemini_api_colab.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "## Image generation and multimodal prompt examples for Gemini 2.0 Flash native image generation"
      ],
      "metadata": {
        "id": "awTm__zIemBw"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "This notebook contains usage examples of the Gemini API for image generation with Gemini 2.0 Flash shared on Talking to Chatbots: https://talkingtochatbots.com/needs-vs-wants-work-life-thoughts-and-gemini-2-0-flash/\n",
        "\n",
        "More info on Google for Developers website: https://developers.googleblog.com/en/experiment-with-gemini-20-flash-native-image-generation/\n",
        "\n",
        "Before running this notebook, follow these steps:\n",
        "\n",
        "- 1. Get your Gemini API Key on Google AI Studio:\n",
        "\n",
        "![Configuring your Gemini API Key](https://talkingtochatbots.com/wp-content/uploads/colab-notebooks/get-api-key-on-google-ai-studio.png)\n",
        "- 2. Activate your Gemini API Key on this Colab environment:\n",
        "\n",
        "![Configuring your Gemini API Key](https://talkingtochatbots.com/wp-content/uploads/colab-notebooks/configuring-your-gemini-api-key.png)\n"
      ],
      "metadata": {
        "id": "kY5DkiQ4fJPm"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Loading API key and importing libraries"
      ],
      "metadata": {
        "id": "l63H7cMTj2il"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BPY4MZJvedaQ",
        "outputId": "1a7dfee7-bfc4-4413-e391-039a57013003"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "API key: AIz*********************************mrA\n"
          ]
        }
      ],
      "source": [
        "# Importing libraries:\n",
        "from google import genai\n",
        "from google.genai import types\n",
        "from PIL import Image\n",
        "from io import BytesIO\n",
        "import matplotlib.pyplot as plt\n",
        "import base64\n",
        "import urllib.request\n",
        "from IPython.display import display, HTML, clear_output\n",
        "\n",
        "# Setting up and checking the API key:\n",
        "\n",
        "COLAB = True # Set this to True for running on Colab\n",
        "USE_DOTENV = False # Set this to False if you don't have a .env file for storing environment variables\n",
        "\n",
        "if COLAB:\n",
        "    USE_DOTENV = False\n",
        "    dotenv_path = None\n",
        "    from google.colab import userdata\n",
        "    gemini_token = userdata.get('GEMINI_TOKEN')\n",
        "\n",
        "if not COLAB:\n",
        "  import os\n",
        "  from dotenv import load_dotenv\n",
        "  load_dotenv(\"../../apis/.env\")\n",
        "  gemini_token = os.getenv(\"GEMINI_API_KEY\")\n",
        "\n",
        "client = genai.Client(api_key=gemini_token)\n",
        "def mask_token(token, unmasked_chars=3):\n",
        "    return token[:unmasked_chars] + '*' * (len(token) - unmasked_chars*2) + token[-unmasked_chars:]\n",
        "print(f\"API key: {mask_token(gemini_token)}\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Gy2mqNUJedaR"
      },
      "source": [
        "## Zero-shot text-to-image"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nQduOKEwedaR"
      },
      "source": [
        "### Needs vs Wants - no template"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "id": "V54PY_ycedaR"
      },
      "outputs": [],
      "source": [
        "def generate_image(prompt, file_name, save_dir=\".\", temperature=0.5):\n",
        "    response = client.models.generate_content(\n",
        "        model=\"gemini-2.0-flash-exp-image-generation\",\n",
        "        contents=prompt,\n",
        "        config=types.GenerateContentConfig(\n",
        "            response_modalities=['Text', 'Image'],\n",
        "            temperature=temperature,\n",
        "        )\n",
        "    )\n",
        "    for part in response.candidates[0].content.parts:\n",
        "      if part.text is not None:\n",
        "        print(part.text)\n",
        "      elif part.inline_data is not None:\n",
        "        image = Image.open(BytesIO((part.inline_data.data)))\n",
        "        image.save(f'{save_dir}/{file_name}.png')\n",
        "    return response, image"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 817
        },
        "id": "m_x4qIpFedaR",
        "outputId": "b11678cc-2366-46e3-ee5f-18fbdff157ee"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<PIL.PngImagePlugin.PngImageFile image mode=RGB size=1024x1024>"
            ],