Will @Kearm20's home GPU setup pass the MMLU-Pro benchmark on the first try?
8
1kṀ13k
resolved Jan 6
Resolved
NO

if this market resolves YES following the first run of the model against the benchmark MMLU-Pro. They need to get within 7% of the announced benchmark result in the DeepSeek v3 paper, as per the market description (if that changes or I described it wrong, the precise criterion is that this linked market resolves YES after the first benchmark run)

  • Update 2025-04-01 (PST) (AI summary of creator comment): - Resolution Conditions:

    • If the other market resolves no, this market resolves no.

    • If the other market resolves yes, this market resolves yes only if the evaluation was completed and the market was resolved yes on the first run.

    • Evaluation Timing:

    • The evaluation must be completed before January 5 to resolve the main market.

    • Evaluations completed after January 5 will not be sufficient to resolve this market.

Get
Ṁ1,000
to start trading!

🏅 Top traders

#NameTotal profit
1Ṁ607
2Ṁ529
3Ṁ225
4Ṁ38
5Ṁ0
© Manifold Markets, Inc.TermsPrivacy