Models: 10 · Tiers: 26 · Samples: 80 · Invalid: 2

Day highlights

GPT-6 Astra

8.67 · max (best)
Model page

medium

8.01 ± 0.20
Judge scores vary between passes

GPT-6.1 Sol

8.35 · max (best)
Model page

low

8.02 ± 0.54
Judge scores vary between passes

GPT-6 Sol

7.88 · max (best)
Model page

low

5.45 ± 1.17
Judge scores vary between passes

GPT-5.6 Sol

7.66 · max (best)
Model page

low

7.34 ± 1.00

medium

7.19 ± 1.17
Judge scores vary between passes

GPT-6 Luna

7.49 · max (best)
Model page

Grok 4.6

7.40 · high (best)
Model page

Grok 4.5

6.75 · high (best)
Model page

GPT-5.6 Terra

6.63 · medium (best)
Model page

GPT-5.6 Luna

5.87 · max (best)
Model page

medium

5.79 ± 1.26

Grok 4.7

5.52 · xhigh (best)
Model page