Chatbot Arena Leaderboard or iterations/new versions of it as creator sees fit. If the project is altogether abandoned with no equivalent, the answers that have not yet resolved YES will resolve NO at EOY unless some replacement appears.
If multiple shared #1 ranks, as has been the case in the past, any of those positions count.
If there are no longer ranked numbers, only top spot/ELO would count. If same ELO, but second spot for some trivial/irrelevant reason, still resolves YES.
Answer resolves YES immediately when an update has been confirmed, no matter what duration it stays in top.
Any AI that hasn't been #1 before during 2025, resolves to NO at the end of the year.
I will not trade in this market, resolution might get messy through mergers/acquisitions, changes to the leaderboard, etc.
Let me know if you want to see improvements to criteria.
See same market but for top 10 below:
/HenriThunberg/who-will-ever-rank-top-10-in-lmsys
Update 2025-02-06 (PST) (AI summary of creator comment): Clarification Added:
The resolution criteria now explicitly include scenarios where OpenAI retakes the lead after January 2025.
Such an event will count toward a confirmed update and will trigger an immediate resolution under the market rules.
Update 2025-07-10 (PST) (AI summary of creator comment): - The "Default" view of the Chatbot Arena Leaderboard will be used for resolution.
The "Remove style control" view will not count.
1,000People are also trading
@PaulHabermas ooops apologies, I forgot myself that I wasn't supposed to trade in this market. Got tempted by your juicy limit order and acted too quickly haha.
Do you feel comfortable with me having a large position against you (hoping that ruling will be easy, and barring that counting on me to be fair) or would you want to set up a limit order for me to sell it back?
@PaulHabermas I think "Default" should in this case be the well... default. I.e. "Remove style control" option is not what counts.
Thanks for your understanding, good luck!
@HenriThunberg resolves true already lol, with ChatGPT-4o latest now rank 1
@HenriThunberg this was a yes on Feb 14 https://x.com/lmarena_ai/status/1890477460380348916
I saw some people are betting it down so might be worth resolving it?
@HenriThunberg Would have to be listed as Microsoft under the "Organization" column of the leaderboard. Same for Apple.