One of the most delightful features of DALL-E 2 is the element of surprise. I most often have no idea how it'll interpret strange queries or concepts I'm trying to convey. A few prompts that surprised and delighted me: labs.openai.com/s/fV3RhC0desjV…
Because I can't spell, I tried this one twice. labs.openai.com/s/wWhzC09TtoFi… I loved all the variations DALLE produced for this prompt, though few looked like the real deal, they all looked delicious.
labs.openai.com/s/UZ3IlhlNRag7… this one had me in tears. Fally Ipupa pandas!
More, pure, hilarious madness: labs.openai.com/s/u1bTQbTCf5NH…
Dall-E had no idea what to do with this query so it gave me great art instead: labs.openai.com/s/chjCShD9R4PF…
Same prompt: labs.openai.com/s/UxexUFtWdbre…
Of course, it'd be impossible to relish the wonders of this technology without considering its risks and limitations, thoughtfully and meticulously outlined here by my incredible colleagues: github.com/openai/dalle-2…. Have a read; we look forward to tackling these challenges.
DALL-E graciously followed my misspelling of Patrice's name. With the right spelling, we get something more coherent: labs.openai.com/s/xKLczNNJq5hR…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
