MANIFOLD
First model series to cross 1500 on lmarena.ai?
49
Ṁ750Ṁ14k
resolved Feb 17
100%99.0%
Claude 4.6
0.0%
Gemini 2.5
0.0%
ChatGPT 4o
0.0%
o3
0.1%
Gemini 3.0
0.0%
Claude 4
0.0%
Claude 4.5
0.0%
Grok 3.5
0.0%
GPT 5
0.0%
o4
0.0%
Deepseek r1
0.0%
Deepseek r2
0.0%
Deepseek v3
0.0%
Deepseek v4
0.0%
GPT-4.1 or similar
0.0%
Claude 4.1 or similar
0.0%
Gemini 2.7 or similar
0.0%
Grok 4
0.0%
Grok 4.1
0.1%
Gemini 3.5

Exactly 1 options resolves yes (unless same exact release date, then split evenly)

(or similar) means anything that is in between 4.0 and 4.5 (for 4.1) or between 2.5 and 3.0 (for 2.7)

includes mini or pro versions

Will resolve if a model stays at or above 1500 for a week and has a 95% CI with a lower bound of at least 1495 at the end of that week (somewhat arbitrary criteria to ensure the score is based on a sufficient amount of votes)

Will N/A if they change the scoring significantly so that a current model passes 1500.

Source:

https://lmarena.ai/leaderboard/text

Default (with Style Control)

Current (6/5/25):

Market context
Get
Ṁ1,000
to start trading!

🏅 Top traders

#TraderTotal profit
1Ṁ216
2Ṁ209
3Ṁ134
4Ṁ83
5Ṁ54
Sort by:

Ok Opus 4.6 over 1500, if it stays there for a week (2/16)

bought Ṁ50 NO

@ChinmayTheMathGuy

"Will resolve if a model stays at or above 1500 for a week and has a 95% CI with a lower bound of at least 1495 at the end of that week"

The lower bound is currently 1494...

@ChaosIsALadder as long as it gets more votes, the CI should tighten. I doubt this will prevent a yes resolution if it actually stays there for a week, but good to point out

@DavidHiggs

Ok it’s 1496 lower bound as of 2/11

boughtṀ262 NO

How can we be that certain 5.2 won’t cross 1500?

I saw the webdev score and was trying to extrapolate

But wasn’t that certain,

Grok 4 only 1433 probably too many Hitler references

grok 4.2 pls

© Manifold Markets, Inc.TermsPrivacy