Just days after OpenAI launched Dall-E 3, its latest and finest AI picture generator, Adobe dropped a brand-new model of its personal AI mannequin known as Firefly 2.0.
So, which one is healthier?
In this text, we’ll discover what’s new in Firefly 2.0 and examine the pictures with Dall-E 3 utilizing comparable prompts.
What is Adobe Firefly?
Adobe Firefly is an AI-powered net utility that may generate photos primarily based in your textual content enter. It works equally to different widespread picture mills like OpenAI’s Dall-E3 and MidJourney.
What units Firefly aside from its rivals is that it’s designed to be protected for industrial use; it’s educated on open-source photos and content material from Adobe Stock.
What’s new in Firefly 2.0?
Firefly 2.0 is a big enchancment over the earlier mannequin. It has a bigger mannequin capability and may generate higher-quality photos, particularly for individuals. It can be sooner and simpler to make use of.
Let’s illustrate these enhancements with an instance:
Prompt: Minimalistic portrait of a younger blonde lady in a room, expressionless, 60s coiffure, 60s vogue, sunrays, mild teal and amber, mushy hues
From this instance, I can see that Firefly 2.0 has enormously improved when it comes to shadows, lighting, pores and skin texture, and particulars. Another vital factor I seen is that 2.0 is quicker in comparison with the older mannequin.
Dall-E 3 vs Firefly 2.0
With Dall-E 3 nonetheless making waves, I used to be interested by its stand in opposition to Firefly 2.0. Let’s put them head-to-head with these examples:
Example 1: cinematic closeup {photograph} of an elephant on the floor of mars, sporting an area helmet
