Models: 10 · Tiers: 26 · Samples: 80 · Invalid: 2
Day highlights
GPT-6 Astra
8.67 · max (best)
Model page max
8.67 ± 0.17medium
8.01 ± 0.20low
7.56 ± 1.11GPT-6.1 Sol
8.35 · max (best)
Model page max
8.35 ± 0.61low
8.02 ± 0.54medium
7.92 ± 0.72GPT-6 Sol
7.88 · max (best)
Model page max
7.88 ± 1.17medium
7.00 ± 0.86low
5.45 ± 1.17GPT-5.6 Sol
7.66 · max (best)
Model page max
7.66 ± 0.67low
7.34 ± 1.00medium
7.19 ± 1.17GPT-6 Luna
7.49 · max (best)
Model page max
7.49 ± 0.51low
5.42 ± 0.38medium
5.31 ± 0.50Grok 4.6
7.40 · high (best)
Model page high
7.40 ± 0.19Grok 4.5
6.75 · high (best)
Model page high
6.75 ± 0.06GPT-5.6 Terra
6.63 · medium (best)
Model page medium
6.63 ± 0.76low
6.35 ± 1.52max
5.33 ± 0.71GPT-5.6 Luna
5.87 · max (best)
Model page max
5.87 ± 1.99medium
5.79 ± 1.26low
4.89 ± 0.58Grok 4.7
5.52 · xhigh (best)
Model page