Changelog
Follow up on the latest improvements and updates.
RSS
new
AI Generator
Added Multiple Image References
Added Multiple Image References to AI Generator's Advanced tab. Available in all paid plans.
This feature allows you to use up to 3 image references at the same time.
new
improved
AI Generator
Added new Image References for SDXL models
Added new Image References for SDXL models:
- Pose
- Soft Edges
- Normal Map
- new version of Depth
- new version of Hard Edges
new
AI Generator
Added option to adjust the number of columns in AI Generator
Added option to adjust the number of columns in AI Generator.
new
AI Generator
AI Generator UI update
We have made significant changes to the AI Generator's Essential tab UI. Our main goal was to simplify and make it more intuitive.
In the Advanced tab, the strength value is now displayed as a percentage. In the "Image to Image" reference, the strength slider now operates inversely to align with the behavior in the Essential tab.
Our plan is to implement these changes to Advanced tab as well.
new
improved
AI Generator
Added 16:9 aspect ratio
We’ve added a long-requested 16:9 and 9:16 aspect ratio to the AI Generator.
We've added 7 new SDXL-based models to our API:
- RealVisXL V4.0
- JuggernautXL X
- Reproduction v3.31
- RealCartoonXL v6
- SDVN7-NijiStyleXLXL
- CounterfeitXL v2.5
- Animagine v3.1
new
AI Generator
Added prompt enhancement
We've added prompt enhancement to AI Generator's advanced tab. It uses AI to generate long, descriptive prompts from short prompts.
new
AI Generator
4 new SDXL models
We’ve added 4 new Stable Diffusion XL-based models to the AI Generator:
- Juggernaut XL X
- RealVisXL V4.0
- ICBINP XL v6
- Animagine XL v3.1
- Added 3 IP Adapters for SDXL: Content, Style and Face --> Documentation
new
AI Generator
Stable Diffusion 3
We’re introducing SD3 to all paid plans.
What’s New with SD3?
- Typography: Robust results in typography.
- Improved performance: Greatly improved performance in multi-subject prompts, image quality, and spelling abilities compared to previous Stable Diffusion models.
- Smart prompt understanding: Delivers high-quality outputs, even for complex prompts and spatial relationships.
Load More
→