Generate an Image
Create images from text prompts using the images.generate method.
require "openai"
client = OpenAI::Client.new
prompt = "An astronaut lounging in a tropical resort in space, pixel art"
# Generate an image based on the prompt
response = client.images.generate(prompt: prompt)
# Prints response containing a URL link to image
pp(response)
Generate with Parameters
Customize image generation with model, size, and quantity parameters.
response = client.images.generate(
model: "gpt-image-1",
prompt: "A cute baby sea otter",
n: 1,
size: "1024x1024"
)
pp(response)
Edit an Image
Edit existing images by providing an image file and a prompt describing the desired changes.
require "pathname"
image = OpenAI::FilePart.new(Pathname('dog.jpg'), content_type: 'image/jpeg')
edited = client.images.edit(
prompt: "make this image look like a painting",
model: "gpt-image-1",
size: '1024x1024',
image: image
)
puts(edited.data.first)
The OpenAI::FilePart class allows you to control the filename and content type when uploading images.
Stream Image Generation
Receive partial images during generation for a progressive loading experience.
require "base64"
stream = client.images.generate_stream_raw(
model: "gpt-image-1",
prompt: "A cute baby sea otter",
n: 1,
size: "1024x1024",
partial_images: 3
)
stream.each do |event|
case event
when OpenAI::Models::ImageGenPartialImageEvent
puts("Partial image #{event.partial_image_index + 1}/3 received")
puts("Size: #{event.b64_json.length} characters (base64)")
# Save partial image to file
filename = "partial_#{event.partial_image_index + 1}.png"
image_data = Base64.decode64(event.b64_json)
File.write(filename, image_data)
puts("Saved to: #{File.expand_path(filename)}")
when OpenAI::Models::ImageGenCompletedEvent
puts("\n✅ Final image completed!")
puts("Size: #{event.b64_json.length} characters (base64)")
# Save final image to file
filename = "final_image.png"
image_data = Base64.decode64(event.b64_json)
File.write(filename, image_data)
puts("Saved to: #{File.expand_path(filename)}")
end
end
The streaming API provides both partial images during generation and the final completed image, allowing you to show progressive updates to users.
# Request base64-encoded response
response = client.images.generate(
prompt: "A sunset over mountains",
response_format: "b64_json"
)
# Decode and save the image
image_data = Base64.decode64(response.data.first.b64_json)
File.write("sunset.png", image_data)