There's so many different versions of each of these models that I think the only reasonable resolution at this point is N/A.

Just extended the closing time to wait for definitive evidence.

Gemini-Ultra does, not betting cause I don't know if it counts

https://storage.googleapis.com/deepmind-media/gemini/gemini_1_report.pdf

@colorednoise Where do you see that?

@Shump ^^^ (same question), also Gemini-Ultra will count for this market

@VictorLJZ It doesn't say it directly, but it does say it beats GPT4 in practically everything, so it seems like a reasonable extrapolation if the results in the paper are real

@colorednoise this will not count for resolution, im waiting for concrete results

By the end of the year, per the close date?

@Joshua no the close date is largely irrelevant here, the results will be determined once we have concrete proof, even if it happens next year