AI Image Generation Gets a Major GPT Image 2.5 Upgrade

The big change to AI Image Generation in GoHighLevel is the arrival of GPT Image 2.5 with two model options: Flare and Sunburst. Instead of using one model for every job, you can now choose an option that better matches the type of image you’re trying to create.

Flare is the default choice for most AI Image Generation tasks. According to the GHL update, it delivers higher quality than GPT Image 2 while offering up to 50% lower latency. That makes Flare a strong fit when you need to generate images quickly or produce a larger number of visuals through a workflow.

Think social media content, product visuals, quick creative concepts, visual search, and rapid prototyping. If your workflow needs to produce images regularly without adding unnecessary wait time, Flare is designed for that kind of everyday use.

Sunburst takes a different approach. It’s built for premium visual work where tighter control across edits matters more. That makes it better suited to polished product imagery, campaign creative, and situations where you’re creating multiple versions of an important visual.

The good part is that both options live inside the existing GHL AI Image Generation workflow action. You don’t need a separate workflow for each model. Open the model dropdown, select Flare or Sunburst based on the job, add your prompt, and continue building your automation.

Better AI Image Generation isn’t only about creating a nicer-looking picture. One of the biggest changes with GPT Image 2.5 is how it handles reference images and keeps important visual details consistent between generations.

When you provide a reference photo, GPT Image 2.5 is designed to keep the subject recognizable across different settings, styles, and compositions. Distinctive features can carry across the generated images while lighting and textures remain more natural. That’s useful when you don’t want every new image to feel like you’re starting from scratch.

This also makes AI Image Generation more practical for creating variations. Instead of rebuilding an entire piece of creative, you can change a specific element while keeping other parts of the image intact. For example, you could swap a product, change the background, or update copy while maintaining the subject, overall composition, and brand treatment.

For GHL users, that’s where this update starts to get interesting. You’re not limited to generating one standalone image. You can build workflows around creating and re-versioning visual assets for different campaigns, audiences, locations, or offers.

That can help turn AI Image Generation into a more repeatable creative process. Rather than manually recreating similar assets every time something changes, GPT Image 2.5 gives you more control over what stays consistent and what gets updated.

AI Image Generation becomes much more useful when it’s part of an automated process. Instead of jumping between GHL, an image tool, and your design workflow, you can generate visual content as an action inside a GoHighLevel workflow.

For agencies, that can remove repetitive creative work. A workflow could generate images for different campaigns, products, offers, or client locations without requiring someone to manually create every variation. Flare is especially useful when speed and volume are the priority.

Sunburst gives you another option when the creative needs more attention. If you’re producing campaign assets or polished product imagery where consistency matters, you can choose a model designed for tighter control across edits.

The improved handling of reference images also makes AI Image Generation more practical. Keeping a subject recognizable while changing backgrounds, styles, copy, or other elements can help you produce related creative without losing the look and feel of the original.

That’s the bigger opportunity for GHL users. Workflows have traditionally handled tasks like messages, lead management, and follow-ups. With stronger AI Image Generation built into those workflows, visual content can become another part of the automation process.

You’re not simply generating an AI image. You’re building a system that can create the right visual as part of a larger marketing workflow.

Setting up AI Image Generation with GPT Image 2.5 happens inside a GoHighLevel workflow. You’ll first access your Workflows page, open the workflow you want to use, and then add the AI Image Generation action at the point where you want GHL to create an image.

Once the action is open, you can choose between Flare and Sunburst, enter the instructions for the image you want to create, and save the action. Flare is the default choice for most image generation, while Sunburst is designed for visual work where you want tighter control across edits.

Here are the steps you’ll follow:

• Access the Workflows page in GoHighLevel
• Open the Workflow Builder
• Add the AI Image Generation action
• Select the GPT Image 2.5 model
• Configure and save the AI Image Generation action

To start make sure you are logged in to your GoHighLevel sub-account.

The Main Menu on the left side of your screen includes the main areas you work in when using GHL.

1.1 Click Automation in the Main Menu.

• Inside the Automation section, you can access your workflows and other workflow-related tools.

1.2 Click Workflows in the top menu if the Workflow List page is not already showing.

• The Workflow List page displays the workflows available inside your sub-account.
• From here, you can open an existing workflow or create a workflow for your AI Image Generation automation.

1.3 Click the workflow where you want to add AI Image Generation.

• The selected workflow opens in the Workflow Builder.
• If you’re building a new automation instead, create the workflow you want to use and open it in the Workflow Builder.

The Workflow Builder shows the trigger and actions that make up your automation. Add the new image action at the point where you want GHL to generate the visual.

2.1 Click the + icon at the position where you want to add the new action.

• The workflow action selection section opens.

2.2 Find and select AI Image Generation.

• The AI Image Generation action settings open so you can configure how the image will be generated.

You can now choose which GPT Image 2.5 variant you want the workflow to use. Your choice will depend on whether you’re prioritizing faster everyday generation or tighter control over the finished creative.

3.1 Click the Model dropdown.

• The available AI image models will appear.

3.2 Select Flare or Sunburst.

• Select Flare for most everyday AI Image Generation tasks, including social content, product visuals, quick prototypes, and higher-volume image creation.
• Select Sunburst when you’re creating more polished campaign or product imagery and need tighter control across edits.

Once you’ve selected your model, complete the remaining AI Image Generation settings based on the image you want the workflow to create.

4.1 Enter the prompt for the image you want to generate.

• Describe the subject and the visual result you want GHL to create.
• Add enough detail to give the model clear direction for the image.

4.2 Complete any additional AI Image Generation settings required for your workflow.

• Review your configuration before saving the action.

4.3 Click Save.

• GHL saves the AI Image Generation action to your workflow.
• Review and test the workflow before using it with live contacts or campaigns.

Your workflow is now set up to use GPT Image 2.5 for AI Image Generation. For most workflows, start with Flare. When you need tighter control over edits or more polished visual output, test the same use case with Sunburst and compare the results.

Getting better results from AI Image Generation starts with choosing the right model for the job. Flare and Sunburst give you different options, so you don’t need to approach every image workflow the same way.

Start with Flare for most everyday image generation. It’s the default GPT Image 2.5 option and is designed for creator content, social images, product visuals, quick prototyping, visual search, and higher-volume generation. GHL also reports up to 50% lower latency compared with GPT Image 2, making it the logical starting point when speed matters.

Use Sunburst when control matters more than volume. It’s designed for premium visual work, including polished product imagery and campaign creative where you may need to make changes while keeping other parts of the image consistent.

Reference images are especially useful when you want consistency between generations. Give GPT Image 2.5 a clear reference when the same person, product, or visual subject needs to remain recognizable across different settings, styles, or compositions.

You can also use AI Image Generation to create variations instead of completely new creative. Try changing one element at a time, such as the background, product, or copy, while keeping the subject, composition, and overall brand treatment consistent.

Finally, test your workflow before putting it into regular use. Compare Flare and Sunburst with the same type of creative so you can see which model gives you the right balance of speed, consistency, and control for that particular workflow.

AI Image Generation becomes more valuable when you stop looking at it as a tool for creating one image at a time. Inside GHL workflows, it can become part of a repeatable process for producing visual content when your business needs it.

For agencies, that could mean creating different versions of campaign creative for multiple clients, locations, products, or offers. Instead of manually starting each image from scratch, you can build AI Image Generation into the workflow that already supports the campaign.

The improvements to reference images create another useful opportunity. You can work from an existing subject and create variations while keeping important visual details recognizable. For example, you may want to change a background, product, or copy while keeping the subject, composition, and overall brand treatment consistent.

Choosing between Flare and Sunburst also gives you more control over how you build these workflows. Flare can handle faster, higher-volume creative tasks, while Sunburst can be used when polished visuals and tighter control across edits are more important.

This doesn’t mean every design task should suddenly become automated. The bigger benefit is having another repetitive marketing task that GHL can help handle inside a workflow. When used for the right campaigns, AI Image Generation can reduce manual creative steps and make it easier to produce visual variations at scale.

The biggest result you can expect from this AI Image Generation update is more choice over how images are created inside your GHL workflows. You can use Flare when speed and volume matter, then switch to Sunburst when the job calls for more control across edits.

With Flare, GHL reports higher image quality than GPT Image 2 with up to 50% lower latency. That can make a noticeable difference in workflows generating social content, product visuals, prototypes, or larger batches of creative where waiting for each image can slow down the process.

You can also expect better consistency when working with reference images. GPT Image 2.5 is designed to keep subjects recognizable across different settings, styles, and compositions. Distinctive features can carry through while lighting and textures remain more natural.

That consistency can be especially helpful when you need variations of an existing creative. Instead of generating something completely different each time, you can change elements such as the background, product, or copy while keeping the subject, composition, and brand treatment anchored.

Your actual results will still depend on your prompt, reference material, workflow setup, and the model you select. Test both Flare and Sunburst with your own use cases before building AI Image Generation into a high-volume workflow. The goal isn’t simply to generate more images. It’s to create useful visuals more consistently with fewer manual steps.

AI Image Generation in GoHighLevel has taken another useful step forward. With GPT Image 2.5 now available inside the workflow action, you have two different models to match the type of creative work your automation needs.

For most workflows, Flare is the place to start. It’s designed for faster generation and everyday tasks like social content, product visuals, prototypes, and higher-volume image creation. When you need polished creative and tighter control across edits, Sunburst gives you another option without moving the process outside GHL.

The improvements to reference images are also important. Being able to keep a subject recognizable while changing individual elements can make AI Image Generation far more practical for campaigns that need multiple versions of the same creative.

The best way to see where this update fits is to test it. Build a simple workflow, generate the same type of image with Flare and Sunburst, and compare the results. Look at generation speed, visual quality, consistency, and how well each model follows your prompt and reference material.

You don’t need to automate every piece of creative. Start with the repetitive image tasks that already take up your time. When the right image can be created as part of the workflow itself, GHL gives you one less manual step between an idea and a finished campaign asset.