I'm impressed by this one. I tried it on audio transcription with timestamps and speaker identification (over a 10 minute MP3) and drawing bounding boxes around creatures in a complex photograph and it did extremely well on both of those.
Plus it drew me a very decent pelican riding a bicycle.
Plus it drew me a very decent pelican riding a bicycle.
Notes here: https://simonwillison.net/2025/Mar/25/gemini/