Grok Build Tops Coding Benchmark

XFreeze · x · 2026-07-10

Grok 4.5 paired with Grok Build took 1st place on the SWE-Atlas-QnA benchmark with a score of 84. The post claims this ties with GPT-5.6 (max) Codex and outperforms other tested configurations like Claude Code Fable 5 (max) and Opus 4.8 (max).

Related event: Grok 4.5 Tops Coding Benchmark with High Token Efficiency(3 posts)→

Original post →

More from coding & agent

coding & agent channel →