ChatGPT 4o models are available in the Invoke hosted application! Start creating with OpenAI's latest groundbreaking image model.
Where do I find ChatGPT 4o models in Invoke?
You can add ChatGPT 4o models to your projects using the 'Model Management' tab in project settings and select the model from the "API" section of the model dropdown in the 'Generation' tab in your Studio.
Additionally, you can access these models using the "ChatGPT 4o - Generate Image" and "ChatGPT 4o - Edit Image" nodes in the Workflow Builder.
What capabilities are available for each model?
ChatGPT 4o has been integrated into Invoke's Canvas and Workflow tools. The main difference between the Low, Medium, and High models is image quality / generation speed and the associated cost.
- ChatGPT 4o - Low: Fastest generation, but at a lower quality.
- Text-to-image
- Global reference images
- Inpainting / outpainting using other models (e.g., FLUX)
- Refactoring / compositing
- Available in custom workflows
- ChatGPT 4o - Medium: Good balance of speed and quality, suitable for most use-cases.
- Text-to-image
- Global reference images
- Inpainting / outpainting using other models (e.g., FLUX)
- Refactoring / compositing
- Available in custom workflows
- ChatGPT 4o - High: Highest quality image with longest generation time.
- Support coming soon
How many credits does a ChatGPT 4o image generation cost?
ChatGPT 4o is an API-based model that will charge a fixed credit amount per request.
Task / Model | Typical time per task* | Typical credits per task* | Capabilities |
ChatGPT 4o - Low | ~10 seconds | ~20 credits | Image generation and editing |
ChatGPT 4o - Medium | ~20 seconds | ~50 credits | Image generation and editing |
ChatGPT 4o - High: | TBD | TBD | Coming soon |
*Actual usage can vary slightly based on image size, inference steps, and advanced settings. The figures above reflect the most common settings our users choose.
You can read more about processing credits and how usage is calculated in our article here.
How is my data handled when I use ChatGPT 4o models?
By default, OpenAI never trains on customer API data, and all image inputs and outputs remain subject to their API usage policies.
Content filtering
The ChatGPT 4o image API leverages the same safety guardrails as 4o image generation in ChatGPT, including safeguards that restrict the model from generating harmful images and the inclusion of C2PA metadata in generated images.
OpenAI will monitor API requests and block requests for generated images that may violate their usage policies, such as child sexual abuse materials and sexual deepfakes. You can read more about creating images in line with their policies here.
Still have questions?
Feel free to contact support or check the release notes for updates as we roll out new models.
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article