Pixel Dojo News
Stay up to date with the latest features and updates from Pixel Dojo.
Introducing MiniMax AI: Next-Level Video Generation
Experience unparalleled video creation with our latest addition: MiniMax AI Video models. Now available for both Image-to-Video and Text-to-Video features, MiniMax brings a new level of realism and quality to your AI-generated videos.
- Highly realistic video generation with smooth and natural movements
- Available for both Image-to-Video and Text-to-Video features
- Impressive handling of complex scenes and multiple subjects
- Create stunning videos from a single image or text prompt
- Seamless integration with existing Pixel Dojo workflows
- Perfect for creating engaging content for social media, marketing, and more
Revolutionary Update: AI-Powered Video Generation Now Live!
Experience the future of content creation with our groundbreaking Image-to-Video and Text-to-Video features, powered by Kling AI technology. Transform static images into dynamic videos or bring your text descriptions to life with high-quality video generation.
- Image-to-Video: Animate any static image, including those created with Flux Pro 1.1
- Text-to-Video: Generate high-quality videos from text prompts alone
- Seamless integration with your custom LoRA models
- Choose between 5 or 10-second video durations
- Uses your existing Pixel Dojo subscription credits
- Perfect for social media content, digital art, and storytelling
Revolutionize Your AI Art: Introducing the Flux LoRA Training Wizard
Experience unprecedented flexibility and power in LoRA model creation with our new Flux LoRA Training Wizard. From one-image wonders to comprehensive datasets, create custom AI models easier than ever before.
- One-Image LoRA Trainer: Create a high-quality LoRA from just a single image
- Multiple Image Upload: Drag and drop up to 12 images for quick training
- Classic Zip File Upload: Use your curated dataset for comprehensive training
- Advanced Beta Settings: Fine-tune your LoRA with customizable training layers and LoRA Rank
- Lightning-Fast Training: Generate LoRAs in as little as 7 minutes
- Compact Models: Resulting LoRAs are just 580KB, 30% faster in inference
Introducing Flux Character Stylist: Revolutionary AI-Powered Character Creation
Experience the power of PuLID technology with our new Flux Character Stylist. Create consistent character images from a single reference photo, leveraging the advanced capabilities of Flux Dev.
- Create consistent character images from just one reference photo
- 12 visual presets for quick styling, including Realistic Portrait, Anime Style, and more
- Utilizes PuLID (Pure and Lightning ID Customization) technology for high-quality results
- Custom prompting with Pixel Dojo enhance feature for detailed image descriptions
- Perfect for creating avatars, digital art, and rapid prototyping in design projects
HUGE UPDATE: New LoRA Gallery & More!
We've just rolled out a massive update that's going to change the game. Explore our new LoRA Gallery, enhanced Model Details Page, and improved privacy controls.
- New LoRA Gallery: Browse all public LoRA Models shared by Pixel Dojo users
- Model Details Page: Copy LoRA URL, download model, add to 'My Models', view trigger tokens and example images
- Privacy Controls: Toggle your LoRAs between Private and Public
- Quality of Life Improvements: Gallery images open in new tabs, keep your place while scrolling
- Minor design changes and bug fixes
Creative Upscaler Update: Enhanced Visuals and Improved Logic
We've just pushed a significant update to the Creative Upscaler, bringing visual style examples, improved 'My Media' section, and enhanced upscaling logic.
- Visual Style Examples: See examples for all optional styles
- Improved 'My Media' Section: Thumbnails now show the final upscaled/relit version
- Enhanced Upscaling Logic: Automatically loads original prompt and adds preset styles
- Better guidance for the upscaler, especially for Flux images
- Can now improve the quality of Flux Pro images
Unleash Your Creativity: FLUX Model Trainer Now Available
Train custom LoRAs using Flux Dev, the world's most advanced AI image model. Create hyper-realistic, personalized styles with unprecedented quality and control.
- Harness the power of Flux Dev for unparalleled realism in custom styles
- Create LoRAs tailored to your specific artistic vision or brand
- Achieve consistent, high-quality results across multiple generations
- Perfect for artists, designers, and brands seeking a unique AI-powered style
- Seamlessly integrate your custom LoRA into the Flux ecosystem
Introducing Flux Inpainting: Redefine Your Images with AI
Experience the power of AI-driven image editing with our new Flux Inpainting feature. Transform parts of your images or add entirely new elements with unprecedented ease and quality.
- Advanced brush tools with size and blur control for precise masking
- AI-powered prompt-based inpainting using Segment Anything technology
- Seamlessly add or replace elements in your images
- Full undo/redo functionality for effortless experimentation
- Perfect for creative edits, removing unwanted objects, or adding new features to images
Major Update: New Flux Realism Model and LoRA Support
We've rolled out exciting new features including a Flux Realism Model, support for multiple FLUX LoRA models, Image Dojo improvements, and more!
- New Flux Realism Model for next-level image and skin quality
- Support for 6 specialized FLUX LoRA models including Midjourney V6, Anime, Art, Disney, Furry, and Scenery
- Image Dojo now supports realism via new dropdowns
- New 'Subtle Enhancements' preset in Creative Upscaler for gentle upscaling
- Automatic trigger words for easy LoRA model usage
Image-to-Image Transformation Powered by FLUX
Experience a new level of image transformation with our updated image-to-image component, now supercharged by the FLUX diffusion model. See the dramatic improvement in quality and detail!
- Incredible enhancement in image quality and detail
- Powered by the state-of-the-art FLUX diffusion model
- Maintain better fidelity to original image structure
- Create more realistic and coherent transformations
- Ideal for style transfer, image editing, and creative reimagining
Introducing FLUX: Next-Gen AI Image Generation
Experience unparalleled image creation with FLUX, our cutting-edge AI model that pushes the boundaries of visual quality, prompt accuracy, and creative diversity.
- State-of-the-art 12B parameter model for superior image generation
- Hybrid architecture combining multimodal and parallel diffusion transformer blocks
- Flow matching technology for enhanced generative capabilities
- Supports diverse aspect ratios and resolutions up to 2.0 megapixels
- Rotary positional embeddings and parallel attention layers for improved performance and efficiency
Introducing: Image Dojo
Revolutionize your AI art creation with Image Dojo, our newest feature that makes generating stunning images easier than ever before.
- Simple natural language input - no complex prompt engineering required
- AI-powered prompt enhancement for high-quality results
- Effortless refinement through conversational feedback
- One-click upscaling for incredibly detailed images
- Perfect for both beginners and experienced AI artists
Create AI Images of ANYTHING in 7 Minutes with LoRA Training
Discover how to rapidly train custom AI models to generate images of any concept - demonstrated with the unreleased Rivian R3X!
- Learn to train object LoRAs in just 7 minutes using PixelDojo.ai
- See how we visualized the unreleased Rivian R3X concept car
- Explore techniques for relighting and upscaling to create stunning wallpapers
- Unlock the potential to generate AI images of virtually anything
New: Creator Profiles & Gallery Pages
We've just rolled out a major update that gives you more ways to showcase your work and connect with others!
- Customizable Creator Profiles: Set up your profile with links to your website and social media.
- Enhanced Image Displays: Your profile appears alongside your images in the Community Gallery.
- Personal Gallery Pages: Get a dedicated gallery link to share your Pixel Dojo portfolio.
Introducing: Kolors
Experience the next evolution in AI image generation with Kolors. This large-scale text-to-image model excels in visual quality, complex semantics, and multilingual text rendering for both Chinese and English.
Try Kolors NowNew: SDXL Style Trainer
Create your own custom AI styles! Our new Style Trainer lets you teach AI your unique artistic vision.
Try Style Trainer NowNew: Aura Flow
Aura Flow is a state-of-the-art AI model capable of generating images from text prompts with high accuracy and detail. Experience the future of AI-powered image creation at Pixel Dojo.
Try Aura Flow Now