Skip to content
assaynotes
EN日本語中文
Evaluation sources

WeirdML

WeirdML / Coding · Implement machine learning solutions on unfamiliar data

Official evaluation
In NewsObserving
Evidence budgetNot scored
Upstream data as ofTo be confirmed
Last syncedNot collected yet

What it measures, and how

Compares on the actual task list and fixed GPU time, keeping the five-round scoring protocol.

How this evidence is used

Under observation: the official CSV is readable, but the score licence and per-model run times have not been fully verified.

Limits and data attribution

Under observation: the official CSV is readable, but the score licence and per-model run times have not been fully verified.

Data licence: Leaderboard data usage boundaries to be confirmed

Scores published by WeirdML. Raw scores and the News consensus score use different scales and cannot be added directly.