Will GPT be the no.1 LLM on this day?
This market will resolve base on the ranking on Chatbot Arena Leaderboard
https://huggingface.co/spaces/lmsys/chatbot-arena-leaderboard
(updated leaderboard link:)
https://chat.lmsys.org/?leaderboard
Resolves Yes for the day if any versions of Chatgpt (which would include GPT3.5, GPT4, GPT4.5, GPT5) is number 1 at any time of the day. (i.e. a screenshot at anytime of the day is sufficient for the day to resolve yes.)
If no one posted a screenshot for Chatgpt being #1 for the day, I will check the ranking on the next day and resolve base on whether I see Chatgpt being #1 on the next day. If you want to make sure you win your bets, post a screenshot!
Clarification:
Number 1 means having the highest Arena Elo. If there's two AI with the exact same Arena Elo, I would count both as number 1.
The leftmost rank column is not relevant for this market
1,000
1.00
@AmmonLam can you do retroactive screenshotless resolutions of 27th and 28th please? :)
@traders Traders, be aware: there was an update earlier today, and now Claude is number 1 on the leaderboard!
@AmmonLam to clarify, by "number 1" do you mean "in the highest spot" or do you mean "has a 1 in the rank column"? LMSYS has started giving "rank 1" to multiple LLMs if the confidence intervals overlap.
@IsaacCarruthers I mean number 1 as in the highest spot, corresponding to highest Arena Elo.
if there's two AI with the exact same Arena Elo, I would count both as number 1
rank is not relevant for this market
@AmmonLam FWIW I definitely had the impression that you meant the # given by hugging face. Rereading your market and description that seems like it was a reasonable interpretation.