Skip to main content
TypingMind has two plugins for OpenAI’s GPT Image models. Both let the AI generate new images and edit images you attach, right inside the chat, using your own OpenAI API key. They differ in which models and output formats they support.
GPT Image 2.6 and GPT Image 2.0 plugins in the TypingMind Plugins store

​
Which GPT Image plugin should I use?

Recommendation: use GPT Image 2.6 for new setups. Both plugins can be installed at the same time.

​
Set up GPT Image 2.6

  1. Click Plugins in the left sidebar.
  2. Under Store → Plugins, find GPT Image 2.6 and click Install.
  3. Open the plugin and go to the Settings tab.
  4. Paste your OpenAI API Key (required). You can create one at platform.openai.com/account/api-keys.
  5. Optionally change the settings below, then click Save.
  6. Make sure the toggle next to the plugin name is on.
GPT Image 2.6 plugin settings: OpenAI API key, model, resolution, quality and SVG model

​
Output formats

The AI picks the format for each request: PNG (default), JPEG, WebP, GIF or SVG. GIF and SVG need a GPT Image 2.6 model.
  • Animated GIF: describe the motion and how it should loop, for example “a looping GIF of a paper boat rocking on waves”.
  • SVG: set SVG Model in the plugin settings to a model ID your OpenAI account can use for SVG. Leave it empty if SVG model selection isn’t enabled for your account.

​
Set up GPT Image 2.0 (GPT Image Editor)

In the store this plugin is listed as GPT Image 2.0. Its overview tab calls it GPT Image Editor.
  1. Under Plugins → Store → Plugins, find GPT Image 2.0 and click Install.
  2. In the Settings tab, paste your OpenAI API Key and click Save.
  3. Turn the plugin on.
GPT Image 2.0 (GPT Image Editor) plugin settings: OpenAI API key, model, resolution and quality

​
Generate and edit images in chat

With a GPT Image plugin turned on, ask in plain language. Generate a new image
Generate a picture of a cat
GPT Image 2.6 generating an image of an orange cat in a TypingMind chat
Edit an image Attach an image, then describe the change:
[Image Attachment] Add a hat to this cat
Editing the generated image in chat: adding a red knitted hat to the cat with GPT Image
You can also edit an image the AI just generated in the same chat. There is no need to attach it again. Keep refining in follow-up messages, for example “make the hat red” or “change the background to a beach”.
Higher quality and larger images use more OpenAI credits. Use medium while you iterate, then switch to high or above for the final image.

​
Troubleshooting

  • No image is generated: open the plugin’s Settings tab and check that your OpenAI API key is saved.
  • Model error: make sure your OpenAI account has access to the selected model, or switch back to the default.
  • GIF or SVG not created: switch Model to a GPT Image 2.6 model.
  • The AI answers in text instead of calling the plugin: check the plugin is on for this chat, then ask explicitly, for example “Use GPT Image to…”.

​
FAQ

The plugins are free to install in TypingMind. Image generation is billed by OpenAI to the API key you add in the plugin settings.
Yes. If both are on, ask for the plugin by name so the AI uses the one you want.
DALL-E 3 was deprecated by OpenAI and the plugin has been removed from TypingMind. Use a GPT Image plugin to generate and edit images instead.