OpenAI's GPT-6 Astra Hits 87.4% on 20-Language Coding Benchmark, Up from 38%
OpenAI has significantly improved its latest models for low-resource programming languages like COBOL, Fortran, and Visual Basic. On the new MultiLangBench, GPT-6 Astra achieves an 87.4% test pass rate across 20 languages, compared to just 38% for GPT-5.6 Terra. For COBOL migration specifically, GPT-6.1 Sol reaches up to 97.9% accuracy, a result that could matter for modernizing critical financial infrastructure. This jump from 38% to 87.4% suggests AI is finally becoming viable for legacy codebases that power banks and government systems.
breakthroughs10/20/2026