Updates | Ideogram 2.0, Model Switcher & More...
✨ NEW MODEL: IDEOGRAM 2.0
The team behind Ideogram 1.0 have quickly followed it up with an even better model - Ideogram 2.0 - and it might just be the best text-to-image model available anywhere right now.
Version 2.0 is even better than the original at spelling and rendering text, and can also handle more complex prompts and generate more detailed images.
Ideogram 2.0 (and it's cheaper counterpart, Ideogram 2.0 Turbo) are available to PRO users, but everyone gets 5 free PRO generations to try it out.
✨ DEFAULT MODEL SWITCHER
Flux will be the new default model for PRO users or anyone with free PRO generations available. Dreamshaper Lightning will continue to be the default model for free users.
Prefer a different model over Flux or Dreamshaper Lightning? No problem, this update also gives you the ability to set your own default model. You'll see a "Set as default" option above the model picker button (as long as you're in advanced mode).
On the topic of Flux - we're working on our Flux architecture to (hopefully) be able to reduce the price soon, plus start adding new features to Flux like Control Nets, LoRA, etc. Stay tuned.
✨ AND SO MUCH MORE:
Image Prompting! 🗣️
"IP Adapter": Copy Subject or Styles (or Both) with Image Prompts.
Use images as prompts to transfer subject matter, style, and composition to new generations.
Image Prompts work differently to normal start images. The model does it's best to understand what's in the image prompt, and then uses that understanding to guide the generation.
Sneak Peek: Home Feed 📰
We're working on a new feed of highly personalised content. It's now available for PRO users to preview.
Want to see what's in store? If you're a PRO user, have a look and let us know if you've got any feedback.
New Lightning Models ⚡️
Access two new powerful SDXL models, retrained from the ground up for realistic and cinematic images (PRO users only). Juggernaut XL v9 Lightning and Juggernaut XI Lightning.
In Case you Missed Them:
✨ New Model: Imagen
Imagen is a next-generation model with incredible prompt-adherence and impressive text rendering capabilities. It comes in two variants: Imagen v3.0 and a faster & cheaper version called Imagen v3.0 Fast.
✨ New Model: Flux
Like Stable Diffusion, Flux is open-source and freely available. Flux outperforms Stable Diffusion in many areas including prompt understanding, text rendering and human anatomy like hands.
✨ Selfie Generator
Simply upload 2-4 images of your face, then enter a text prompt (or choose from one of our presets) to generate an image of yourself as an Astronaut, cowboy, Van Gogh painting, sci-fi protagonist, or anything else you can imagine!
✨ New Model: Ideogram
Ideogram is a powerful AI text-to-image model that can can generate text in images better than any other model.
✨ Animate Your Creations
Turn any creation into a short video with just a few clicks. Get a new perspective, watch your creations spring to life, or just experiment to see what happens!
✨ New Model: DALL·E 3
The latest from OpenAI represents a leap forward in the ability to generate images that exactly adhere to the text you provide, understanding significantly more nuance and detail than previous systems.