You cannot create any edited images with this new “GPTs”. Where is the claim that you can? As usual, the strict guidelines make the product useless, not to mention the user helps
I can only add to this. The description of “Hot Mods” literally says “Let’s modify your image into something really wild.”, but when prompted to do so the response is "I’m unable to manipulate images directly or create modifications to existing images. " and at best it creates some useless generic image.
Seems to me like the premise of Hot Mods is a baseless claim.
This Hot Mods GTP is pretty good at gaslighting me, it just keeps telling me, its coming together, give me a few more minutes… Hours later, more promises, but never ever does it deliver. Quite funny really!
Hi, welcome to the community!
I explained it following topic, but OpenAI fixed only Data Analyst, not others.
That GPT’s capability (DALL-E) is disable.
In the first image, you can see that the DALL-E tool is enabled, while the other two are disabled.
Interestingly, second and third are OpenAI’s own GPTs.
You can see here. First one is working because its DALL-E tool is enabled, but other two are not working, but they say they created images 
If you have Plus subscription you can create a custom GPT using following prompt:
You are a visual enhancement assistant. Your role is to help users creatively visualize modifications or decorative additions to their existing images. Your output must preserve the original image’s structural integrity, core composition, and color palette while offering visually compelling, high-concept enhancements. These enhancements should be imaginative, stylish, and refined, ranging from subtle refinements to bold artistic reinterpretations, without distorting the essence or primary subjects of the image. Maintain a balance between artistic creativity and conceptual coherence. Avoid gimmicks; focus on tasteful, innovative suggestions that elevate the visual impact of the original content.
Or you can find many similar GPTs on Explore GPTs.
And following GPT is one of them. I used the prompt above I provided.
I just built it after your post, very fresh, just cooked 
https://chatgpt.com/g/g-67e84b8f89708191b0378ae81379a867-visual-refine
It also is nothing special, a placeholder up there with “Laundry Buddy”. The complete context:
The GPT will assist users in visualizing modifications or decorations to their images. It will maintain the image basic integrity and color while providing creative visual enhancements. Be very creative, but preserve high concepts
The GPT and yours also gets hit with the standard ChatGPT “vibe-matching” prompt:
Over the course of the conversation, you adapt to the user’s tone and preference. Try to match the user’s vibe, tone, and generally how they are speaking. You want the conversation to feel natural. You engage in authentic conversation by responding to the information provided, asking relevant questions, and showing genuine curiosity. If natural, continue the conversation with casual conversation.
Here’s negations of that default and realignment in a GPT. Should not be fooled into “getting back to you” or other tonal annoyances (but is only as good as the AI model). It also has DALL-E without any special “hot mods” instructions.
OpenAI is also dumping info about the image being “a legacy model” into AI context to further confuddle the output after an image generation.
“Hot Mods” is a GPT made by OpenAI back when GPTs were first introduced. It was just a cute little prompt of instructions. You cannot enable features on other’s GPTs.
It indeed is getting no image creation tool, neither old nor new. The GPT will hallucinate up to where it has no tool to call. You can simply ask in a normal chat though, with the unstoppable “image_gen” tool, to make some sweet mods to your input image, thus this GPT is obsolete anyway.
They are not my GPTs, but OpenAI’s own GPTs.