Changelog
Follow up on the latest improvements and updates.
RSS
new
improved
Image Generator
Image Editor
Gemini 2.5 Flash Image added to Image Generator and Image Editor
- Image Generator–Gemini 2.5 Flash Image is now available as a new model, delivering fast, high-quality image generation with improved control and detail.
- Image Editor– Text editing now uses Gemini 2.5 Flash Image, offering exceptional consistency across edits—subjects remain recognizable even after style or appearance changes.
improved
Video Generator
Added last frame support to Kling 2.1 Pro
Kling 2.1 Pro now supports both first and last frames, giving you greater control over the video creation process.
Available in Video Generator.
improved
Image Generator
Updates in Image Generator
- Qwen Image in free plan– now available without subscription.
- Image References for Qwen Image– added Qwen Edit Reference to guide image generation with a reference image.
- Improved prompt helpers– Random Prompt and AI Improve tuned for better text handling and more diverse outputs.
Available in Image Generator
improved
Image Editor
Updated Text Editing in Image Editor
Text editing now uses Qwen Image Edit instead of FLUX Kontext to produce sharper, cleaner edits with better layout awareness.
Available now in the Image Editor.
new
Image Generator
New image generation model: Qwen Image
Qwen Image is now available in the Image Generator. It produces detailed compositions with vivid colors and supports a wide range of styles — from photorealism to artistic concepts. Works well with complex scenes and visual storytelling.
new
Image Generator
Added 5 new public styles to FLUX.1 [dev]
We’ve added 5 new public styles to the Image Generator. They are ready to use in FLUX.1 [dev] and are available to all users, including those on the free plan.
New styles:
- Tentacle Twist
- Tintype Wetplate
- Victorian Surreal
- Vintage Film
- Watercolor
Styles can be applied directly without training and help give generated images a distinct visual look.
new
improved
Video Generator
Added image-to-video mode to Google Veo 3 and Veo 3 Fast
Image to Video mode is now supported in both Google Veo 3 and Google Veo 3 Fast. In addition to generating videos from text prompts, you can now start your video using an uploaded image as the first frame.
Both models continue to offer high-quality motion, visual consistency, and built-in audio — now with more control over how your videos begin.
Available now in the Video Generator.
new
Video Generator
New video generation model: Wan 2.2
Wan 2.2 is now available in the Video Generator, with support for both image-to-video and text-to-video. It offers great motion quality, visual richness, and precise prompt responsiveness.
The model focuses on expressive camera movement and visual control — producing coherent, cinematic results even in complex scenes.
new
Image Generator
Added 5 new public styles to FLUX.1 [dev]
We’ve added 5 new public styles to the Image Generator. They are ready to use in FLUX.1 [dev] and are available to all users, including those on the free plan.
New styles:
- Sepia Glitch
- Slide Grunge
- Solar Tapestry
- Stained-Glass
- Sumi-e Ink
Styles can be applied directly without training and help give generated images a distinct visual look.
new
Image Generator
Added 5 new public styles to FLUX.1 [dev]
We’ve added 5 new public styles to the Image Generator. They are ready to use in FLUX.1 [dev] and are available to all users, including those on the free plan.
New styles:
- Neon Melt
- Neon Wabi-Sabi
- Pinhole Blur
- Pink Doll World
- Rembrandt Mood
Styles can be applied directly without training and help give generated images a distinct visual look.
Load More
→