Gpt Image 1
4 modes on this page
Gpt Image 1 turns a text prompt into an image. Pick a mode, write the prompt, run it in the studio.
Input
FormMode
Prompt*
Image Size
Number of Images
Quality
Background
Result
Preview
Illustrative preview, not Gpt Image 1 output
$0.002 per image·~10s
What Gpt Image 1 is for
Gpt Image 1 from OpenAI is a text-to-image model. You describe the frame and it returns a still image in about 10 seconds, billed at $0.002 per image. This page groups 4 modes (Text to Image, Edit Images, Text to Image, and Edit) so you are not hunting separate URLs for the same family.
Gpt Image 1 is cheap enough that a generation usually fits the free Knowlify plan's credits. A typical run is about 10 seconds. Published rate: Your request will cost different amounts based on the input and output. - You will be charged $0.002 per 1,000 input text tokens. One word is roughly 4 tokens. - For low quality, you will be charged $0.011 for 1024x1024 or $0.016 for any other size *per output image*. - For medium quality, you will be charged $0.042 for 1024x1024 or $0.063 for any other size *per output image*. - For high quality, you will be charged $0.167 for 1024x1024 or $0.25 for any other size *per output image*. - Your total request cost will be rounded up to the nearest cent.
Pick Gpt Image 1 for text to image at $0.002 per image and ~10s. Compare it with Gpt Image 1.5, Gpt Image 2 on the related cards below if you need a different look, speed, or input type.
How to use Gpt Image 1
- 1. Write a still prompt for Gpt Image 1. Be specific about subject, style, and lens.
- 2. Run it. The prompt opens in the Knowlify studio, where Gpt Image 1 generates.
- 3. Stay in the studio to turn a document or a longer prompt into a narrated video.
What you need
- A prompt that names the subject, style, and framing.
- Image Size (auto, 1024x1024, 1536x1024, 1024x1536).
- Quality (auto, low, medium, high).
Example prompt
a lion walking down an empty midnight boulevard, fashion film still, anamorphic, teal and gold, no text, no watermark
Frequently asked questions
How do I run Gpt Image 1?
Run Gpt Image 1 on Knowlify at /generate/gpt-image-1. This page has 4 modes in one dropdown. Write the prompt on this page and it opens in the studio, where Gpt Image 1 generates a still image. The same studio also turns documents into full narrated videos.
Where can I use Gpt Image 1 online?
On Knowlify (OpenAI's model, knowlify.com/generate/gpt-image-1). You do not need a separate playground. Gpt Image 1 runs in the studio next to document-to-video.
Can I make a longer video with Gpt Image 1?
Gpt Image 1 itself produces a still image. On Knowlify you can take that output, or skip it and upload a document, and generate a full narrated video with a script you can edit.
What do I need to run Gpt Image 1?
A text prompt is enough. Gpt Image 1 generates the image from that description in the studio.
How long does Gpt Image 1 take?
Plan on about 10 seconds for a typical run, then whatever queue time the studio has. Video models sit at the high end of that range.
Does Gpt Image 1 have more than one mode?
Yes. This page has 4 modes: Text to Image, Edit Images, Text to Image, and Edit. Switch them in the Mode dropdown instead of opening a new URL.
What does gpt-image-1 cost?
Your request will cost different amounts based on the input and output. - You will be charged $0.002 per 1,000 input text tokens. One word is roughly 4 tokens. - For low quality, you will be charged $0.011 for 1024x1024 or $0.016 for any other size *per output image*. - For medium quality, you will be charged $0.042 for 1024x1024 or $0.063 for any other size *per output image*. - For high quality, you will be charged $0.167 for 1024x1024 or $0.25 for any other size *per output image*. - Your total request cost will be rounded up to the nearest cent. Knowlify charges this as credits, at the compute cost.
What is gpt-image-1 best for?
OpenAI's latest image generation and editing model: gpt-1-image. It turns a text prompt into an image.
Do I need an account to use gpt-image-1?
A free Knowlify account, yes. This page composes the prompt; the studio is where gpt-image-1 actually runs.