Add materials for "How to Generate Images With the OpenAI API"

generate-images-openai/README.md

@@ -0,0 +1,44 @@
+# How to Generate Images With the OpenAI API
+
+Learn to use the OpenAI Python library to create images with the GPT Image models. In the associated tutorial on [generating images with the OpenAI API](https://realpython.com/generate-images-openai/), you'll create images from text prompts, tune the size and quality of your results, and edit an image with a follow-up prompt. You'll learn how to interact with the Images API and incorporate this functionality into your Python scripts.
+
+## Setup
+
+Create and activate a virtual environment, then install the `openai` package:
+
+```console
+$ python --version
+Python 3.14.6
+$ python -m venv venv
+$ source venv/bin/activate
+(venv) $ python -m pip install openai
+```
+
+You need to be on Python 3.10 or higher.
+
+The scripts read your API key from the `OPENAI_API_KEY` environment variable:
+
+```console
+(venv) $ export OPENAI_API_KEY="<your-key-value-here>"
+```
+
+## Create and Edit Images
+
+Follow the instructions in [the tutorial](https://realpython.com/generate-images-openai/) to create images from text prompts, edit an image with a follow-up prompt, and convert a Base64 JSON response to a PNG image file.
+
+You can find the code for each of these steps in dedicated scripts:
+
+- `create_png.py`: Create an image from a text prompt and write it straight to a PNG file.
+- `create.py`: Create an image from a text prompt and save the whole JSON response to a file.
+- `convert.py`: Convert a Base64-encoded PNG image delivered in a JSON response to a PNG image file.
+- `edit.py`: Read Base64-encoded image data and make an API request to edit that image with a prompt.
+
+In the tutorial, you'll walk through each of these scripts and their functionality and output in more detail.
+
+Note that `convert.py` and `edit.py` read a saved response from the `responses/` directory. Run `create.py` first, then update the `JSON_FILE` and `SOURCE_FILE` constants to match the filename that it wrote.
+
+## A Note on Models
+
+The GPT Image models replaced DALL·E, which OpenAI [retired on May 12, 2026](https://developers.openai.com/api/docs/deprecations) along with the image variations endpoint. These scripts call `gpt-image-2`.
+
+OpenAI retires image models on a regular schedule, so if a script starts failing with a message like `The model 'gpt-image-2' does not exist`, then check the deprecations page and swap the model name.

generate-images-openai/convert.py

@@ -0,0 +1,18 @@
+import json
+from base64 import b64decode
+from pathlib import Path
+
+DATA_DIR = Path.cwd() / "responses"
+JSON_FILE = DATA_DIR / "An ec-1786009967.json"
+IMAGE_DIR = Path.cwd() / "images" / JSON_FILE.stem
+
+IMAGE_DIR.mkdir(parents=True, exist_ok=True)
+
+with open(JSON_FILE, mode="r", encoding="utf-8") as file:
+    response = json.load(file)
+
+for index, image_dict in enumerate(response["data"]):
+    image_data = b64decode(image_dict["b64_json"])
+    image_file = IMAGE_DIR / f"{JSON_FILE.stem}-{index}.png"
+    with open(image_file, mode="wb") as png:
+        png.write(image_data)

generate-images-openai/create.py

@@ -0,0 +1,24 @@
+import json
+from pathlib import Path
+
+from openai import OpenAI
+
+client = OpenAI()
+
+PROMPT = "An eco-friendly computer from the 90s in the style of vaporwave"
+DATA_DIR = Path.cwd() / "responses"
+
+DATA_DIR.mkdir(exist_ok=True)
+
+response = client.images.generate(
+    model="gpt-image-2",
+    prompt=PROMPT,
+    n=1,
+    size="1024x1024",
+    quality="low",
+)
+
+file_name = DATA_DIR / f"{PROMPT[:5]}-{response.created}.json"
+
+with open(file_name, mode="w", encoding="utf-8") as file:
+    json.dump(response.to_dict(), file)

generate-images-openai/create_png.py

@@ -0,0 +1,18 @@
+from base64 import b64decode
+
+from openai import OpenAI
+
+client = OpenAI()
+
+PROMPT = "A vaporwave computer"
+
+response = client.images.generate(
+    model="gpt-image-2",
+    prompt=PROMPT,
+    n=1,
+    size="1024x1024",
+    quality="low",
+)
+
+with open("vaporwave.png", mode="wb") as png:
+    png.write(b64decode(response.data[0].b64_json))

generate-images-openai/edit.py

@@ -0,0 +1,29 @@
+import json
+from base64 import b64decode
+from pathlib import Path
+
+from openai import OpenAI
+
+client = OpenAI()
+
+DATA_DIR = Path.cwd() / "responses"
+SOURCE_FILE = DATA_DIR / "An ec-1786009967.json"
+EDIT_PROMPT = "Add a large potted plant growing out of the computer"
+
+with open(SOURCE_FILE, mode="r", encoding="utf-8") as json_file:
+    saved_response = json.load(json_file)
+    image_data = b64decode(saved_response["data"][0]["b64_json"])
+
+response = client.images.edit(
+    model="gpt-image-2",
+    image=("image.png", image_data),
+    prompt=EDIT_PROMPT,
+    n=3,
+    size="1024x1024",
+    quality="low",
+)
+
+new_file_name = f"edit-{SOURCE_FILE.stem[:5]}-{response.created}.json"
+
+with open(DATA_DIR / new_file_name, mode="w", encoding="utf-8") as file:
+    json.dump(response.to_dict(), file)

generate-images-openai/requirements.txt

@@ -0,0 +1,16 @@
+annotated-types==0.8.0
+anyio==4.14.2
+certifi==2026.7.22
+distro==1.9.0
+h11==0.16.0
+httpcore==1.0.9
+httpx==0.28.1
+idna==3.18
+jiter==0.16.0
+openai==2.53.0
+pydantic==2.13.4
+pydantic-core==2.46.4
+sniffio==1.3.1
+tqdm==4.70.0
+typing-extensions==4.16.0
+typing-inspection==0.4.2