PREVIEW · the site is new and still changing
Nerdict
← ModelsOpenAI · gpt-6-luna · runs in Codex CLI 0.162.0GPT-6 Luna

79/80 checks · $0.028 · 7m50s for 5 tests, at high thinking

In Video 1

What it built

high thinking · middle of 3 runs

By thinking level

more thinking = slower, usually pricier