Is it possible to use gpt-image-1 in Pickaxe for accurate text-based image generation?

[moira777]

Hi Pickaxe team ,

I’m building a visual assistant on my website that generates images based on user prompts, and I specifically need the ability to create images that contain clear, accurate text (e.g., posters with phrases like “Psychology that transforms lives”).

I’ve tried using both the Flux Shell and the “Use your own OpenAI key” options in Pickaxe, but both seem to rely on dall-e-3, which produces nice visuals but frequently introduces text errors (like duplicated or misshaped letters).

Since OpenAI recently made gpt-image-1 available via their API (as shown in their official docs), I’d love to know:

1. Is it currently possible to use gpt-image-1 for image generation in Pickaxe?
2. If not, are there any plans to support it soon as an alternative to DALL·E 3?
3. Is there any manual way to set the model in custom actions using my API key?

Thanks in advance and congrats on building such a great and flexible platform!

[thomasumstattd]

It is possible now. You can find GPT Image-1 in Actions. You will need to use your own API key.

Here is a post to help:

GPT Image-1 Action | How to add custom image parameters

The create GPT image action now let you use the latest GPT-image-1 API. To further customise your image you can do the following: 1\ Clone the action [image] 2\ Select the action code [image] 3\ Modify the data variable with the required parameters [image] background (optional) Allows to set transparency for the background of the generated image(s). This parameter is only supported for gpt-image-1. Must be one of transparent, opaque or auto (default value). When auto is used, the model…