We made Grok 4.5, GPT-5.5, and Claude build the same appsVisit link →Grok 4.5 just launched. So we had it, GPT-5.5, Claude Opus 4.8, and Fable 5 one-shot the same interactive apps, then measured latency and cost. Here is who won.July 10, 2026ai performancePermalink: 2026/w28/we-made-grok-45-gpt-55-and-claude-build-the-same-apps Copy Related LinksGitHub - peonist-ai/halogen-flash-server ai performanceHarnessTax: How Much Does the Harness Matter for Coding Agents? ai performanceRTK reports huge token savings, but our cost benchmarks disagree - Quesma Blog ai performanceGitHub - redhat-et/ripwire performance aiFast and Hard Code ai performance← Back to Week 28
2026/w28/we-made-grok-45-gpt-55-and-claude-build-the-same-apps