Personal benchmark paragraph — revision

5 modelslatest public run of each

Instructionssee my feedback

see my feedback

Sorted by score. Open a card to read the whole answer beside its checks.

GPT-5.6

21 Sep · Codex CLI · high effort

4/5 (80%)
Loading…
4 of 5 checksOpen →

GPT-6 Sol

22 Sep · Codex CLI · high effort

4/5 (80%)
Loading…
4 of 5 checksOpen →

GPT-6.1 Sol

1 Oct · Codex CLI · high effort

4/5 (80%)
Loading…
4 of 5 checksOpen →

Sonnet 5.5

24 Sep · Claude Code · high effort

4/5 (80%)
Loading…
4 of 5 checksOpen →

Opus 5.5

19 Sep · Claude Code · high effort

4/5 (80%)
Loading…
4 of 5 checksOpen →