3.0.0
2023-10-03
New realtime feed showcasing all the best diffusions from our users.
More models and more LoRAs!
2.8.1
2023-07-11
Base- and LoRA-model pickers now more compact.
When repeatedly diffusing, will now default to using the current diffused image as the next initial image.
This pattern is now also reflected accurately on screen - previously would continue to show wrong initial image.
Prompt boxes default to using height set by user. The previous "auto-hug" functionality can be enabled in Settings.
Many small user experience (UX) and performance improvements.
2.8.0
2023-07-04
New models!
The introduction of LoRAs! (they're cool)
Inpainting-specific models (for better results)
Resizable prompt boxes
Profile picture will now stick around across reinstalls
Long-press on Info to copy said info to your device's clipboard
More robust networking. Won't swoon and give up as easily now
Many small UI/UX enhancements to make things just a little better here and there
A few too many bug fixes
2.7.1
2023-06-13
A couple small bug fixes.
2.7.0
2023-06-12
Inpainting is back! It's not entirely New and Improved but it's back!
Better stability and performance with Diffusion History.
Can now properly specify prompts for inpainting. No more faces within faces!
User can choose between original and diffused image for subsequent use/diffuse actions.
Info now shows all of the information.
Black Screen of Death no longer added to history.
Will now save diffused image rather than initial image.
2.6.1
2023-06-02
Bug fixes:
Daily totals now more accurate
Saving an image will save the diffused image rather than the initial image.
Better feedback handling
2.6.0
2023-05-30
Added Realistic Vision model. Create ultra-realistic images!
Unlock unlimited Diffusions with one low-price subscription
Diffusion and Prompt histories
Stability and speed improvements
2.5.0
2023-05-02
Subscriptions!
And a new model to play with.
Lots and lots of small improvements and bug fixes.
2.4.1
2023-04-16
Enhancements:
We now disable the phone's idle timer only while actively diffusing. Otherwise, your phone can sleep again.
Added more help sections to explain why things happen.
App now limits image caches to 100 MB.
Bug Fixes:
Three day NSFW grace period now works better.
Users can now successfully delete their own accounts.
Password recovery now works as expected.
Resolved issue where CFG scale was getting "stuck".
Seed value now shows up in diffusion history cards.
2.4.0
2023-04-10
New Features
* Registration! Ok, but it's needed to enable certain kinds of features.
* You can now set a profile image which might be handy for some social features coming soon.
* Diffusion History is now available so you'll never accidentally lose your art again!
Enhancements
* As an added safety measure for our users, registration is now required to view NSFW content.
* You can now choose to blur NSFW content. You know, if you want to.
* We've added cool new Deliberate-based images to the Gallery.
* You can turn off swipe-to-dismiss on temporary (modal) screens
* You can now toggle brush sizes in the in-painting screen!
* Buttons are now a little bigger. No more finger sniping.
* Can now redo brush strokes while in-painting.
* Content loads faster!
Bug Fixes
* App now accurately reports image sizes (go figure).
2.3.4
2023-03-07
Quick Save now much more responsive and provides visual feedback.
Now easier to dismiss temporary (modal) screens.
Doing image-to-image with initial image set to 0% will now no longer include the original image on the result screen.
Fixed bug where model gallery thumbnail wasn't being displayed properly.
Fixed edge case where the wrong model would be used.
Now adds watermark to auto-saved images (if watermarks is enabled).
2.3.3
2023-03-06
Fixed bug where some users were getting blurry images
Fixed bug related to the help button not being displayed in initial image area
Fixed bug where model gallery thumbnail wasn't being displayed properly
Fixed bug that sometimes caused a crash with certain types of images
Various performance enhancements. Not JUST bug fixes!
2.3.2
2023-03-01
Fixed bug that caused the initial image to be saved rather than the diffused image.
Fixed bug that didn't always clear out previous in-painting brush strokes.
Sometimes a swipe would accidentally dismiss a modal screen. Now it should do what users intend.
If there was an initial image and the user put the app in the background, the image will always be restored now.
Added Discord link to the Feedback screen so there can be a two-way conversation with users, if they want.
Added Auto Save. If this setting is enabled the app will automatically save ALL diffusions to the camera roll.
2.3.1
2023-02-27
Fixed bug when importing images from galleries
A "What's New" screen (how meta)
New delta view so users can easily compare before/after images when using initial images
More help area and little UX tweaks to make things better/cleaner/easier
2.3.0
2023-02-20
Better performance, both in the app and in the cloud, improved inpainting, improved image-to-image, support for iOS 15 again, added "quick save" functionality, all-around better user-experience for generating images, can now cancel networking requests, bug fixes
2.2.1
2023-01-29
Added ability to remove watermark since all current users should be able to do it
Reduced watermark to just “DIFFUSITRON” and made it about 10% smaller
Fixed in-painting bug.
2.2.0
2023-01-26
New models!!!
1) We upgraded the anime model (Anything) to v4
2) We added an awesome new model called "Deliberate". Check it out!
Also... some bug fixes and that kind of stuff.
There's a link to our Discord in Settings, come join us!
2.1.0
2023-01-23
Now supports in-painting!
Increased performance
Bug fixes (hardly any of those, of course)
2.0.2
2023-01-03
Tweaks to the UI that should make some things easier
User can now have image generation temporarily halted when they lose wifi, thus avoiding costly cellular charges (thanks for the suggestion, random user!)
A couple small -- super small -- bug fixes
2.0.0
2023-01-02
Curated gallery of images and "workflows" to help you get started.
Generating images from existing images. Turn yourself into a celebrity or see what you look like as an anime character.
Multiple style support. We've curated a few styles for you to try, such as anime and ultra-realistic people. With more to come later.
More settings to allow you to tailor the app to work exactly how you want it to.
1.17.0
2022-11-29
New onboarding and walkthrough experiences
Better performance and more consistent UI
A few bug fixes, not many. :)