1,000๐ Top traders
| # | Name | Total profit |
|---|---|---|
| 1 | แน424 | |
| 2 | แน393 | |
| 3 | แน382 | |
| 4 | แน271 | |
| 5 | แน193 |
People are also trading
Dunno how to see past versions of the leaderboard, but this market had the same criteria and resolved to google: https://manifold.markets/Soli/who-will-have-the-best-llm-by-end-o
Creator (@AbhinavSrinivas) seems to be inactive. I'm wondering what should happen if multiple companies share rank 1, since that is now possible on LMArena. My reading is that the company with the highest arena score (Elo) should win, just like this question. Do others read it this way as well?
Similar markets w/ arbitrage opportunities:
https://manifold.markets/Soli/who-will-have-the-best-llm-at-the-e
https://manifold.markets/Soli/who-will-have-the-best-llm-by-end-o
How will this market resolve if Meta trains the best LLM, open sources it, and then Mistral finetunes it for a small to moderate amount of additional tokens which makes it perform somewhat better than the original Meta model at the end of 2024?
(As seems to be what's going on with mistral-medium.)
@RyanGreenblatt I think Iโll run a poll to see what people want. Or maybe Iโll resolve both as yes.
@mattyb Good question. Iโm mostly thinking using this leaderboard: https://chat.lmsys.org/ (go to the leaderboard tab). If you have any other suggestions then let me know.