Prome AI is a tool that promises to generate stylish images and videos from varying inputs. It claims outputs of high quality worthy of use in professional and creative spaces. However, in a world plagued by numerous tools that fail to deliver on their promises, Prome AI was put to the test.
In this article, we will explore the features of PromeAI and its abilities as an image and video generation tool.
Main Features
1. Image Generation
PromeAI can generate images from text prompts. It uses generative AI models to interpret text and develop it into a rich image bearing all the prompt details within seconds. PromeAI gets its strength of image richness from its vast repository of artistic styles.
Therefore, images can take on a photorealistic, anime, digital painting, line art, 3D, or cyberpunk style.
PromeAI also provides prompt modifiers and style tags for more customization and fine-tuning. Armed with these settings, users can alter details like lighting, composition, color tone, or character posture.
Also read: Ideogram Review
2. Sketch Rendering
PromeAI can turn rough hand-drawn or digital sketches into polished images. To use this feature, the user will be required to select a visual style from the style library. This style forms the reference and will guide the rendering process for the output image.
The resulting image will retain the original structure of the drawing, but the AI fills in details on textures, lighting, and depth based on the reference style. PromeAI can work well with basic outlines, CAD-like technical drawings, and more detailed drafts.
Also read: AI Image Generation from Existing Photos (Tutorial and Tools)
3. Creative Fusion
This Photo-to-Style Fusion feature, also known as Creative Fusion, merges photos with reference images to produce hybrid outputs. During processing, it retains the structural elements (pose, layout, or framing) of the original image while incorporating a new aesthetic.
The intensity and direction of the fusion are controlled by the user. These settings essentially tell the AI how much of the reference style should be applied.
4. Video Generation
PromeAI has a video generation feature that creates short videos from text and images. Using either a still image, a prompt, or both, the AI creates an image with frames that show consistency with the prompt.
This feature can be used to explain products, visualize campaign concepts, or share short stories on social media.
5. Model Switching
PromeAI has three different modes, each with differing capabilities and costs. It allows users to choose between modes based on resources and design needs.
Using Prome AI
Rendering a Sketch
To start, I uploaded a sketched image, typed in a prompt, and added a reference image.
I proceeded to pick the style and went for the ‘Natural’ option.
I further adjusted settings on artistry, which dictates the amount of freedom the tool will have in the rendering. Then I selected a perspective and a type of lighting. I did note the numerous interesting options to choose from.
The results:
The pictures came out with a lot of semblance to the artistry of the reference image. The AI did, in fact, stick to the prompting.
Hovering the cursor over each resulting image prompted information about the style, mode, creativity, and lighting. It also came with options to remix or publish.
Editing an Image
There was also a list of edits that could be made.
I branched off to explore the HD upscaler. Setting the creativity to high, I typed in the same prompt I had used when generating the image and let the AI work its magic. Just before I clicked ‘generate’, I decided to check out the ‘Template’ setting, which had two options: ‘Natural’ and ‘Anime.’ I left it at “Natural.”
I did encounter a slight hiccup just before the processing was completed. The tool stated that the image was either too small or too large for the process, but I could click the ‘continue’ option to get on with it, irrespective. I did just that.
From the result, I noticed no obvious difference between the initial image and the upscaled image. It probably had something to do with the size.
The same editing options were presented. My guess is that the features exist in a loop, and an image can be further edited until a desired upscale outcome is achieved. Though that may take multiple edits to achieve.
Using the Erase and Replace Feature
Taking another route, I used the ‘Erase and Replace’ feature to apply alterations. Once selected, I noted that the AI did a rapid image analysis, probably to identify elements/objects.
There was a list of options here, too. I could replace, retouch, remove, insert an object, or recolor. I decided to remove objects.
Once I moved the cursor across the image, the corresponding highlighted portions (in purple) appeared. What I liked was that the AI was able to identify and section each object/element in the image.
The result wasn’t what I expected. It appeared that the AI chose to perform a form of autofill, using the elements on the computer screen as a reference, as opposed to leaving it blank. 
Image to Video
Next, I decided to explore the image-to-video feature. I resorted to the original, unedited rendered image for this. I attached the prompt “A young man engrossed in a gaming session with his friends. He should occasionally shout reactions like ‘Yeah!’ “ to guide the AI along.
I adjusted the motion intensity to 100 and clicked ‘Generate’
The result was a 4-second-long video that showed a lack of transition between frames. The subject, the young man, in the video, moved. Then, the camera zoomed in on the video game where the character made slight movements about its axis.
Frame 1
Frame 2: A zoom in on the computer screen to show the game character
Attributing the video quality to the motion intensity, I set it back to the default (80) and regenerated the video. The result included slight movements by the subject, which was expected as gaming is not typically motion-intensive.
However, in this particular video, there were no rapid zoom-ins or faster-than-the-eye transitions between frames.
The Bottom Line
PromeAI is an excellent tool for generating and rendering images. Its long edit window allows for creative branching out that’s necessary to make a true masterpiece.