GPT-5.6 Benchmarks and Costs Revealed

iamrobotbear · x · 2026-07-10

Pre-release evaluations from Artificial Analysis show that OpenAI's GPT-5.6 Sol nearly matches Claude Fable 5 on the Artificial Analysis Intelligence Index, but at roughly one-third of the cost.

The same evaluation indicates that GPT-5.6 Terra and Luna scored 55 and 51 respectively, with even lower costs per task; additionally, GPT-5.6 Sol leads the Coding Agent Index within OpenAI's Codex harness.

Related event: GPT-5.6 Series Shines in Benchmarks: Tops Coding and Offers Better Cost-Efficiency(14 posts)→

Original post →

More from coding & agent

coding & agent channel →