Skip to main contentSkip to footer
ExamplescriptintermediateRunnableguided-flow

References

Runnable example (intermediate) for script using openai, pillow.

Key Facts

Level
intermediate
Runtime
Python • OpenAI API
Pattern
Inspectable flow with visible system boundaries
Interaction
Live sandbox • Script
Updated
14 March 2026

Navigate this example

High-level flow

How this example moves from input to execution and reviewable output
References -> Initialize OpenAI client -> Create model response -> Render the visible result -> video

Trigger

References

Runtime

Initialize OpenAI client

Outcome

Create model response

Why this page exists

This example is shown as both real source code and a product-facing interaction pattern so learners can connect implementation, UX, and doctrine without leaving the library.

Visual flowReal sourceSandbox or walkthroughMCP access
How should this example be used in the platform?

Use the sandbox to understand the experience pattern first, then inspect the source to see how the product boundary, model boundary, and doctrine boundary are actually implemented.

UX pattern: Inspectable flow with visible system boundaries
Source references
Library entry
models-openai-08-video-2-references
Source path
content/example-library/sources/models/openai/08-video/2-references.py
Libraries
openai, pillow, pydantic, python-dotenv, requests
Runtime requirements
OPENAI_API_KEY
Related principles

2-references.py

python
import time
from openai import OpenAI
from pathlib import Path
import base64
from utils.resizer import resize_image
from utils.downloader import download_sora_video

openai = OpenAI()


# --------------------------------------------------------------
# Generate a reference image
# --------------------------------------------------------------

# Takes abbout 2 minutes
response = openai.responses.create(
    model="gpt-5",
    input="A professional studio desk setup with Shure SM7B microphone on boom arm. Dark background with soft blue LED light glow creating ambient atmosphere. Clean desk surface, professional lighting equipment visible. Cinematic look, moody lighting, no people.",
    tools=[
        {
            "type": "image_generation",
            "size": "1024x1536",
            "quality": "high",
        }
    ],
)


# --------------------------------------------------------------
# Resize the image to 720x1280 and save it
# --------------------------------------------------------------

message_outputs = [output for output in response.output if output.type == "message"]

if message_outputs:
    for msg in message_outputs:
        for content in msg.content:
            if hasattr(content, "text"):
                print(f"API Response: {content.text}")

image_data = [
    output.result
    for output in response.output
    if output.type == "image_generation_call"
]

if image_data:
    image_base64 = image_data[0]
    image_path = f"./references/{response.id}.png"
    with open(image_path, "wb") as f:
        f.write(base64.b64decode(image_base64))
    resize_image(image_path)
    print(f"Saved and resized image to 720x1280: {image_path}")
else:
    print("No image was generated. Check the API response above.")

# --------------------------------------------------------------
# Use input reference image
# --------------------------------------------------------------

# The moderation here is quite strict.
# You might need to try different images/prompts if you get an error.

video = openai.videos.create(
    model="sora-2",
    prompt="Make this image come to life",
    input_reference=Path(f"./references/{response.id}.png"),
    size="720x1280",
    seconds=4,
)

print("Video generation started:", video)


# --------------------------------------------------------------
# Get last video
# --------------------------------------------------------------

time.sleep(3)
last_video = openai.videos.list().data[0]

# --------------------------------------------------------------
# Download the video
# --------------------------------------------------------------

video = download_sora_video(video=last_video, output_folder="./output")
What should the learner inspect in the code?

Look for the exact place where system scope is bounded: schema definitions, prompt framing, runtime configuration, and the call site that turns user intent into a concrete model or workflow action.

Look for output contracts and validation
Look for the exact execution call
Look for what the product could expose to the user
How does the sandbox relate to the source?

The sandbox should make the UX legible: what the user sees, what the system is deciding, and how the result becomes reviewable. The source then shows how that behavior is actually implemented.

Read the implementation summary.
Step through the user and system states.
Inspect the source code with the highlighted doctrine decisions in mind.
SandboxInspectable flow with visible system boundaries
Interaction walkthrough

Use the sandbox to step through the user-visible experience, the system work behind it, and the doctrine choice the example is making.

UX explanation

The sandbox explains what the user should see, what the system is doing, and where control or inspectability must remain explicit.

AI design explanation

The page turns raw source into a product-facing pattern: what the model is allowed to decide, what the product should expose, and where deterministic code or review should take over.

Interaction walkthrough

  1. 1Read the implementation summary.
  2. 2Step through the user and system states.
  3. 3Inspect the source code with the highlighted doctrine decisions in mind.

Visible to the user

A script demonstrating openai + pillow.

System work

The product prepares a bounded model or workflow task.

Why it matters

The interface should make the delegated task legible before automation happens.

Used in courses and paths

This example currently stands on its own in the library, but it still connects to the principle system and the broader example family.

Related principles

    Runtime architecture

    Use this example in your agents

    This example is also available through the blueprint’s agent-ready layer. Use the For agents page for the public MCP, deterministic exports, and Claude/Cursor setup.

    Define triggers, context, and boundaries before increasing autonomy
    Make control, observability, and recovery explicit in the runtime
    Choose the right operational patterns before delegating to workflows