@thtbee_ 3.5 Flash did the same thing in /teamwork-preview
I believe that is the way the orchestration works with /teamwork as opposed to being model specific
@genevieve__h I love Gemini! I feel tho that the benchmark for this has changed of late. It is not whether the agent can complete, but rather is it willing to find and exploit a 0 day flaw with the booking software and bump people off so that I can get my preferred date and time.
gemini 3.5 flash is sooooooo good at image reasoning
image reasoning is when the model has to actually think about what it sees
not just label objects or read text
it needs to connect the facts and figure out whatโs going on
reasoning trace:
current order top to bottom: 654, 770, 655, 771, 772
target order: 654, 655, 770, 771, 772
LIS is 4: 654, 655, 771, 772
slide 770 between 655 and 771
answer: 1