@Bayesian how will this question be resolved? The next model will likely be 3.5 and I don't think we will get Grok 3 Thinking

@Fynn

i don't even see grok 3 thinking beta in the api, confusion

@Bayesian maybe close and N/A if nobody has a better idea?

why is april not near 0? are we assuming it might actually have been released as grok 3.1 already (per https://x.com/testingcatalog/status/1920806273324331055)?

How does this resolve if it's not released by end of Jan 2026?

@outsideness it's a bit cursed bc it's probable that they're just rebranding grok 3 to grok 3.5 bc of further post training but i think the intent is for jan 2026 to be "jan 2026 or later / never"? I'm not sure

@Bayesian OK, I'll hold you to that

@outsideness Bad idea, i said im not sure. Either it’s that or I N/A the market due to the ambiguity.

@Bayesian well, I think it's how these new date markets are meant to work, just like other numeric markets work