Mobile App Tutorial
Become a PRO subscriber
PRO subscribers currently enjoy unlimited diffusions, inpainting, upscaling, and ControlNet, along with commercialization ownership. Upgrade to a PRO Diffusitron subscriber by navigating to the Admin section located at the bottom right of your screen and clicking on “Manage”.
Prompts
Writing an effective prompt is crucial in AI art, as it directly influences the style, content, and quality of the resulting artwork. Begin by entering your prompt in Studio, using the box labeled “Things you want to see”. Clearly define your objective or the concept you want to convey. Is it a person, thing, theme, mood, or idea? The clearer you are in expressing your intent, the better.
Use concise and focused language, avoiding overly complex instructions. Incorporate descriptions and specific details about style, elements, colors and reference works. You can even specify the intended audience or purpose of the artwork.
Remember to experiment, as AI art is an iterative process, and exploring different prompts will lead to diverse and surprising results. You can even use emojis! Don’t be discouraged - most AI artists don’t get it right the first time. The more you learn about prompt generation and advanced settings, the better you can iterate to get it just right.
PRO TIP: Make sure the most crucial element - the focal point - is written first in your prompt. Provide detailed descriptions, followed by the setting, including location, mood, camera angles, and lighting.
Structuring your prompt using brackets and numbers to assign weights to specific components can guide the AI to emphasize certain aspects of the image. Break down the prompt into individual components that use brackets to assign weights, with additional brackets adding emphasis. Double, triple, quadruple the brackets to add emphasis.
PRO TIP: If you add brackets around everything then nothing is important, as everything is the same and you risk lowering the quality of your work. Use them sparingly.
Incorporating weight numbers within the components of your prompt also communicates their importance. Adding a number between 1.1 and 2.0 at the end of a word or phrase will increase emphasis, while numbers between 0.0 and 0.9 decrease it.
PRO TIP: If you’re unsure where to start, especially with prompt weighting, simply type your concept in the prompt box and click the “Enhance” button in the bottom left corner of the box. This feature optimizes your prompt with a single click!
Keep in mind that while adding brackets or numbers can enhance results, it’s not necessary for producing high-quality art. It is simply another advanced feature at your disposal.
 |
 |
|---|---|
| Example Prompt 1 | Example Prompt 2 |
To manage your prompts, use the X button in the top left corner to delete, and the arrows in the bottom right corner to cycle through previous prompts. If “Enable Prompt Auto Hug” is toggled in your Admin settings, the prompt box will adjust automatically. You can manually resize the boxes by dragging them down using the gray horizontal line, or double-tap the line for automatic resizing, which is ideal for longer edits.
Negative prompts
Negative prompts are written in the Studio box titled “Things you don’t want to see” and involve instructing the AI on what to avoid or what content not to include. This includes characteristics, elements, styles, colors, shapes, textures, or subjects. Similar to positive prompts, specificity and concise language are highly important, as vagueness may lead to unintended results.
Negative prompts can also incorporate brackets and weights in the same way as positive prompts.
PRO TIP: Many experienced AI artists have a master set of negative prompts that they regularly use, tailored to their preferred style of subject matter. They then add a few theme-specific phrases at the beginning of each new artwork. Observing how other artists do this through the Diffusitron community Feed or on Discord is a fantastic way to create your own master list of negative prompts.
 |
 |
|---|---|
| Example Negative Prompt 1 | Example Negative Prompt 2 |
The X in the top left corner of the prompt box will delete your prompt. Additionally, you can cycle through your previous prompts using the arrows in the bottom right corner of the box.
If you have enabled “Prompt Auto Hug” in your Admin settings, the prompt box will automatically resize to fit the inputs as you type. To manually adjust the box size, drag the gray horizontal line beneath the box downward. Double-tapping the line will trigger automatic resizing, which is particularly useful for extensive edits.
Models
When creating an image, selecting the best model is essential. Models are trained on large datasets of existing artworks or creative content, learning to generate new content that mimics the style, structure or characteristics of the training data. Therefore each model excels at only certain tasks while struggling with others. Choosing the right model is crucial for achieving the desired outputs once you have your prompt, but it’s also an iterative process.
PRO TIP: After inputting the prompt, the best thing to do is to try several models or all available models and narrow down the one(s) that best match your desired effect. Then, iterate on the prompt again and run it through the narrowed-down models.
In some cases, merging models can have the best outcomes. Some users use specific models to generate people with particular characteristics, then run the image again through image-to-image with another model to maintain those characteristics while applying different styles. For example, transforming an anime character into a realistic photographic representation. ControlNet also offers similar possibilities. Given that some models excel at adding detail to an image while others focus more on composition, combining models can be highly effective!
 |
 |
|---|---|
| Model: Real Cartoon Anime | Model: EpicRealism |
Models are constantly evolving, with new ones frequently added. By clicking on the four-dotted icon next to the Model box, users can access summaries of all the models. These summaries include brief descriptions of the models’ strengths and five images created using each model to help select the best model for your needs.
LoRAs
While selecting a model is required when creating an image, opting for a LoRA is optional. LoRAs, or Latent Overlapping Representation for Adaptations, are small fine-tuned models that are used in combination with existing models. Simply put, LoRAs enhance the capability of the models. They can introduce new elements to an image, such as characters, art styles, poses, details, and more. By incorporating a LoRA in with your chosen model you can generate greater diversity and variation. The overlapping latent regions enable smooth transitions between different image styles or characteristics, are much easier to use than image-to-image to add these elements, and result in more artistically interesting outputs, all while seamlessly integrating with the model.
 |
 |
|---|---|
| No LoRA | Manga LoRA |
Aspect Ratios
The aspect ratio of an image is the ratio of its width to its height.
The AI will create an image according to the selected aspect ratio UNLESS an initial image is uploaded to image-to-image, inpainting, or ControlNet. Aspect ratio will also not work if you are diffusing directly from an image screen preview, as it uses image-to-image for processing. Diffusitron’s aspect ratios are optimized for the best quality; in some cases, this may involve the AI generating a smaller image, which is then upscaled to the desired aspect ratio. This approach is necessary to maintain image quality, as a direct upscale could result in a lower-quality image based on the AI’s training. The currently supported aspect ratios are as follows:
| Ratio | Image Size in Pixels |
|---|---|
| 1:1 | 1024 x 1024 |
| 3:4 | 1152 x 1536 |
| 9:16 | 1152 x 2048 |
| 4:5 | 1024 x 1280 |
| 4:3 | 1536 x 1152 |
| 16:9 | 2048 x 1152 |
Upscale (Pro Feature)
You can achieve larger images (higher pixel values) by using the upscale button before saving your images to your phone or sending them out into the world. To access the upscaler, click on your image to preview, then navigate to the Tools option in the bottom menu and select Upscale. Note that this is a PRO feature. While the image may appear the same, downloading it to your phone or computer will result in a larger size and improved quality. The new upscaled aspect ratios and pixel values are as follows:
| Ratio | Image Size in Pixels | Upscaled Image Size in Pixels |
|---|---|---|
| 1:1 | 1024 x 1024 | 2048 x 2048 |
| 3:4 | 1152 x 1536 | 4608 x 6144 |
| 9:16 | 1152 x 2048 | 2304 x 4096 |
| 4:5 | 1024 x 1280 | 4096 x 5120 |
| 4:3 | 1536 x 1152 | 6144 x 4608 |
| 16:9 | 2048 x 1152 | 4096 x 2304 |
Steps
STEPS is an acronym that stands for Style Transfer Enhanced by Pyramidal Synthesis, a method used by the AI for style transfer. In other words, it refers to the number of steps the AI takes in generating your image, where the style of one image is applied to another. Increasing the step value requires more computation and prolongs the diffusion time. Currently, the number of steps supported range from 20 to 50.
PRO TIP: Although it is possible to have higher step counts, most images do not improve beyond 30 steps, so the tradeoff in computation time and image quality tends to diminish above 30 in most cases.
CFG Scale
CFG stands for Classifier-free Guidance, which adjusts how much the output will be like your prompt by employing neural networks, particularly in style transfer and image synthesis tasks. Be cautious with this setting as it can have profound undesirable effects on your results, especially at higher values.
PRO TIP: Higher CFG values tend to produce more creative images, while lower values are a better bet if you have specific images and styles in mind. CFG has an effect on the content in terms of the spatial arrangement and semantic information, as well as texture, color, and style. If you are producing consistently odd and low quality images, check your CFG setting first!
CFG scale ranges in the app from 1 to 20.
 |
 |
 |
|---|---|---|
| CFG at 1.0 | CFG at 10.0 | CFG at 20.0 |
Seed Values
Seed Values are the initial input parameters or random values that influence the generation process. You can use the same seed to obtain the same image output as a previously generated image, while using random seeds allows for exploration of different outputs. It is usually best to enable “Use Random Seed” for beginners. If you have a specific AI image you want to recreate, and you have all the other information to do so (such as prompts, negative prompts, CFG, steps, etc.), inputting the seed value can precisely guide the generation process. Seed values serve as a starting point for the generation process and control the creative output of the models.
Image-to-Image
Image-to-image is a way to manipulate and transform images, creating new variations based on the entire initial image provided. It’s useful when you want to retain most elements of the original image, particularly color and composition. Image-to-image maps between input images and output images, and takes into account your prompt and negative prompt. The strength of this mapping depends on the percentage allocated by the user. A higher image-to-image strength percentage results in the image closely resembling the initial upload, with less emphasis on new set parameters, like prompts. At 100% image strength, the image remains identical to the original with no variation. Lower strength will look a lot less like the original image, with more emphasis on the positive and negative prompts for guidance.
The keyboard icon in the top right hand of the image-to-image window allows precise percentage input without using the slider.
Clicking on “Provide Initial Image” opens a menu for selecting your image input. You can import directly from your phone’s camera or gallery, use images from the Diffusitron Feed as templates, or select an image from your own Diffusion History. To replace an initial image with a new one, click the image thumbnail next to the percentage slider to reopen the input menu.
 |
 |
|---|---|
| Original | Image-to-Image |
Inpainting
Inpainting allows you to selectively target and modify specific parts of an image while leaving the rest of the image as is. Note that this is a PRO feature.
In order to “mask” which areas you want changed, start by clicking on the PAINT button. This opens your masking screen, where your image is displayed in preview alongside five options on the bottom: undo, redo, brushes, keep and cancel. You can easily zoom in and out of your image by pinching the screen.
Begin by clicking on the Brushes icon. Here, you can choose from five different brush sizes, ranging from 1 to 50 pixels, to precisely define the areas of your artwork. You can use the undo and redo buttons as needed. If you are satisfied with your choice, click Keep. Please note that the app will not remember your choice if you close the window without clicking Keep. To exit the window without selecting any masking area for inpainting, swipe down, tap the down arrow on top, or click the Cancel button.
PRO TIP: Inpainting requires a defined mask. If inpainting is not working, ensure that a mask has been selected. This is the first step to troubleshooting if you encounter issues, as your mask may not have been saved or may have reset from a previous session. Painting a mask and closing the window to diffuse will not save it. You must select KEEP before closing the window.
Once you have chosen your masking area, specify what the AI should replace the areas with by typing in the prompt box (titled “Fill painted areas with this”) and negative prompt box (titled “Exclude this”).
 |
 |
 |
|---|---|---|
| Configuration | Original Image | Inpainting Output |
Similar to image-to-image, the strength of inpainting depends on the percentage allocated by the user. You can accurately input the percentage using the keyboard icon in the top right corner. A high percentage indicates that the masked area will be completely transformed into a new image, while 100% strength will solely rely on the new prompt and negative prompt.
PRO TIP: Be careful when using a high percentage, as it may result in images lacking cohesion with the original artwork, resembling a poorly-cut collage with sharp demarcation edges.
A low percentage strength will minimally impact the original image, resulting in subtle differences. Inpainting strength varies greatly depending on the artist’s intent. Click DIFFUSE on the bottom of your app to generate the new image.
ControlNet (Pro Feature)
ControlNet has been added to provide users with more structural and artistic control over their images. It utilizes fine-tuned models to condition the input images. For example, some models enable the generation of entirely new images but preserve poses, while others will keep the integrity of the edge detections.
PRO TIP: It is always a good idea to test out several models to determine which is best for your specific need. The choice of model depends on factors such as the type of initial image uploaded and the desired output.
Begin by uploading an initial image by clicking “Provide Initial Image” in the middle of the Studio tab. If an initial image is already present, simply click on the current image to bring up the menu. You can upload an initial image from your phone camera or gallery, Diffusion Feed, or Diffusion History.
Select the ControlNet model you intend to use. Make sure your prompt (“Things you want to see”) and negative prompt (“Things you don’t want to see”) at the top of Studio reflect the image you want the AI to create, as it will be changing your image with those prompts and the model strength you set. In the case of ControlNet, you must choose a ControlNet model, a model from the Model gallery, and you have the option of also choosing a LoRa. A low percent ControlNet strength will minimally impact your initial image. A high percentage strength will have a profound effect on your initial image. You can input the percentage accurately using the keyboard icon in the top right corner.
Note that this is a PRO feature.
 |
 |
 |
|---|---|---|
| Original Image | Controlnet Settings | Control Net Output |
Face Fix (Pro Feature)
If you have created an amazing image but notice a distorted face, you may be able to automatically fix it with just the click of a button! Generally speaking, the face fixer aims to correct or enhance facial features. However, the quality of the face fix will depend on the original image and the AI’s ability to recognize the face accurately.
To try fixing the face automatically, open the “Tools” menu on the bottom of the image and select “Face Fix”. Note that this is a PRO feature. Keep the diffusing image open, as you’ll have the opportunity to compare the fixed image with the original using a slider on the far right of the screen. Your new image will be saved in your Diffuistron History automatically.
User History
Your user history can store up to 200 images and never reaches its maximum capacity. Only the last 200 images are retained at any given time. Deleting images frees up space, so you can be sure to use the 200 maximum effectively.
There are several benefits to storing your best images within the app. Saving them elsewhere does not retain the necessary information to reproduce the image, create variations using tools like image-to-image or ControlNet, reuse positive or negative prompts for other works, or access other information such as Seed Values, models, or LoRAs. Additionally, having the image saved in your favorites allows for easy comparison and upscaling.
Having your image already saved in the app makes it convenient to pick up editing in another session or automatically upload it to the Diffusitron desktop Studio version for more editing features.
You can view your 200-image history in two ways: a carousel at the top of your Studio screen, or in cards with time stamps and detailed information. Note that both methods typically show only a portion of the image in the thumbnail.
In the Studio carousel, pressing and holding a thumbnail allows for quick deletion. Tap once more to leave edit mode. You can mass delete your entire history with a tap of the X in the top left corner of the carousel. Tapping an image will pull it up in preview mode, where you can easily edit, share, rediffuse and view all info.
Tapping the icon with four squares above the carousel takes you to the card view. Scroll thumbnails and the information overview, and swipe left or right to delete an image. Tapping on the top-down arrow will close the card view, and you can swipe down to close if you have “Swipe to Dismiss” enabled in Admin.
Image View Mode
Tapping on an image, whether from your Diffusion History or from the Feed, will open the image in a full view mode, offering you a variety of options.
The Social button allows you to “Share with the rest of the world”. You can save the image to your phone, send the image outside the app via email or text, or post on social media.
Additionally, within the Social button, you can choose to “Publish to the app feed”, which submits the image for approval to the community feed. Note that publishing to the feed also includes the prompts used to make the image, so ensure your name is updated in your profile under Admin for proper credit attribution.
In the “Tools” button you have the ability to compare images, use upscaling, and apply face fix. Please note that upscaling and face fix are PRO features. You also have the option to delete images in the tools box; however, this option is not available when viewing images you didn’t create or upload, such as those from the Feed.
Create a variation of the image by simply tapping “Diffuse!” at the bottom center of the screen. This action automatically applies image-to-image, allowing you to create an image with similarities in composition, color, style, as well as the same size/ratio of the image used. Staying on the screen while diffusing enables you to compare the original and diffused versions using the slider on the far right.
Tapping “Use” at the bottom of the screen returns you to the Studio with the prompt, negative prompt, and image loaded into image-to-image. This feature is useful for tweaking the prompt, negative prompt, customizing the image-to-image percentage, or editing the image without re-entering the information.
PRO TIP: If you hit the “Use” button but the “Use Image From Workflow” toggle is off in your Admin Settings, the image WILL NOT load into image-to-image, though the prompt and negative prompt will load. Check the Admin settings, found in the bottom right of your screen, if you are not getting the expected settings.
Pressing the “Info” button at the bottom of the image view mode provides a detailed summary of the image creation process, including the artist, NSFW classification, the positive and negative prompts used, model and LoRA (if any) used, if ControlNet was used, dimensions, CFG scale, Steps, and Seed value used.
To close the Image View Mode, swipe down. If “Swipe to Dismiss” is disabled in your Admin Settings, tap the down-arrow to dismiss instead.
Publishing Your Work in the Community Feed
Users can contribute images to the community Feed by clicking on the “Social” button in your image view. Choose the option “Publish to the app feed” to submit the image for approval. Note that publishing to the feed will also include the workflow.
Make sure that your name is updated in your profile, found under Admin, so that the images are credited to you! Most AI artists use the same pen name across all social media platforms, on Discord, and within the app. It’s great for branding your work and getting to interact with like-minded people!
While Diffusitron is an open app, we have strict guidelines against any images depicting harm to minors. We maintain stringent filters within the app to prevent such content. If you see an inappropriate image, please don’t hesitate to contact us via feedback (located in the upper right corner next to the Diffusitron title or on Discord). If you believe your image was unfairly flagged, reach out to our team; we’re happy to help!
Feed Images as Templates
The community feed is an excellent way to find inspiration and refine your craft. Experienced AI artists develop their own style, prompts, and workflows. Studying works featured in the community feed provides valuable insights into refining your own creative process. Despite all the amazing advancements in AI art, predicting the outcome is still difficult. There’s always an element of surprise, and achieving a specific result can be challenging. Even with training a model or LoRA yourself, some degree of unpredictability will always be there. Taking the time to study how certain images are made is an excellent way to hone your skills and understand how each setting influences the final output.
You can recreate similar images found in the Feed by tapping “Diffuse!” at the bottom of the image screen. The AI will generate an image using image-to-image that closely resembles the original work. Staying on the screen while it diffuses allows you to compare the original and new versions using the compare slider in the far right corner. Note that NONE of the settings will be loaded into your Studio.
To load all the settings into your Studio, simply press the “Use” button at the bottom of the image screen. This action imports all settings used to create the image, without diffusing the image.
PRO TIP: When clicking the “Use” button, the initial image will be loaded into image-to-image at 50% strength by default, unless you have disabled “Use Image From Workflow” in the Admin settings. In such cases, only the prompt and negative prompt will be loaded into your Studio, without the image. If you encounter difficulties loading images as templates from the Feed, double-check the Admin settings for possible solutions!
Comparing Images
Comparing images is an excellent way to refine your workflow. It enables you to easily identify differences between variations, select higher-quality images, and finetune your prompts.
To begin, open your image and select the “Tools” option at the bottom. From the popup menu, select “Compare”. You can compare your diffused image to images found in your phone camera, phone gallery, the Diffusitron Feed, your personal Diffusitron History within the app, or the initial image used to create your image, if it is a variation. Once selected, simply use the slider on the right side of the image to compare.
 |
 |
|---|---|
| Example 1 | Example 2 |
When comparing images, two gray dots will appear at the top of the view. The selected gray dot determines which image the AI will utilize for any subsequent editing. Choosing the dot in the top left corner selects the top image, while selecting the dot in the top right corner selects the image being compared underneath. You can then download or share the selected image, upscale, face fix, diffuse it again, or use it as a basis for comparison with another image in your workflow.
Admin Preferences
Below is a definitive list of options within the Admin tab. Our Privacy Policy and Terms of Use are easily accessible by scrolling to the bottom of the tab.
Customizing your preferences helps to make Diffusitron a more personalized and enjoyable experience.
- At the top of Admin you can go directly to the Diffusitron Discord. Share your work, ask questions, find inspiration, and find your tribe!
- Customize your user profile by adding a profile image (click on the + icon). Customize your username above your email address; this is the name that will appear publicly when you submit images to the community feed. PRO TIP: Using the same username across the app, Discord, and other social media platforms helps to establish your brand as an artist!
- Enabling “Multi Diffusion Mode” allows you to initiate multiple diffusions simultaneously, rather than running them one at a time and waiting for them to finish. This is particularly useful when you’re confident about all of your settings, or just want to experiment with one setting while keeping the rest as is.
- Toggling “Block NSFW Content” will prevent images classified as not safe for work from being displayed. Such content, including nudity and gore, will be blurred from both the community feed and your own images in your Diffusion History.
- Enabling “Include Watermark'' adds a small Diffusitron watermark to downloaded or shared images. Unlike other apps, we do not require you to watermark your art, but we appreciate you spreading the word!
- Toggling “Enable Prompt Auto Hug” automatically resizes the prompt text boxes as you type. Double-tapping the gray horizontal line under the prompt boxes also automatically resizes them.
- Enabling “Swipe to Dismiss” allows you to swipe away temporary screens. Some people don’t like the swipe feature as it can be easy to accidentally dismiss the screen. If disabled, you can dismiss screens by pressing the down-arrow located at the top of the screen.
- Having the “Enable Auto Save'' mode on automatically saves created content, eliminating the need to manually save one-by-one to your phone’s camera roll. That way you’re ready to go if you are using images and/or variations in bulk for projects.
- “Enable Quick Save” allows you to save individual images by pressing down on them for one second, without having to go through the multi-step “Social” menu. Make sure to hold down until you see the dropdown message! This will work on any screen that also has a Share button.
- Enabling “Low Data Mode” ensures images are only downloaded when on Wi-Fi, preventing cellular data usage.
- Enabling “Diffuse from Keyboard” adds a Go button to the keyboard to start diffusing after editing your prompt or negative prompt. If disabled, the keyboard will have a Done button to dismiss it.
Permanently Delete Your Account
To permanently delete your account and all of your personal data from the app, go to Admin, scroll to the bottom of the page, and click the red Delete button. Make sure you are absolutely certain as all of your work will be lost!
Feedback
Send feedback directly to the developers by clicking on the chat icon next to the DIFFUSITRON title at the top of your screen. You can request an answer back via email. Please note that this feedback is not linked to the App Store reviews. If you wish to leave us a review, please do so through your Apple account. Your feedback is greatly appreciated!