Will Grok 3 beat DeepSeek r1 in LiveBench?
51
1kṀ14k
resolved Apr 11
Resolved
N/A

Resolves as soon as Grok 3 has a rating in https://livebench.ai. DeepSeek-r1 currently has a global average of 71.57, which Grok 3 would have to beat for this market to resolve as YES.

Credit to @ChaosIsALadder for the market format.

  • Update 2025-03-01 (PST) (AI summary of creator comment): Resolution Criteria Update:

    • The market will resolve based on the first global score of a Grok 3 model.

AI
OpenAI
AI Benchmarks
DeepSeek
Grok 3
Get
Ṁ1,000
to start trading!

People are also trading

Sort by:

at this point it seems sort of likely that a very different model will be the first with api access... can we NA this now?

what happens if grok3 is significantly updated before all its scores are released on livebench?

@CrypticQccZ i think it just resolves based on the first global score of a grok 3 model

sold Ṁ32 NO

@jim resolves yes now, in that case

@jim ok grok3 is below deepseek but grok3 mini high thinking is above. imo this should be an NA

Which Grok 3? The reasoning variant or non reasoning model?

People are also trading

© Manifold Markets, Inc.•Terms + Mana-only Terms•Privacy•Rules