OpenAI's GPT-6 models achieve 87.4% pass on MultiLangBench and 97.9% COBOL migration accuracy
OpenAI's GPT-6 Astra scored an 87.4% test pass rate across 20 low‑resource programming languages on the MultiLangBench, up from about 38% for GPT-5.6 Terra. Its GPT-6.1 Sol variant reached up to 97.9% accuracy on COBOL migration tasks. The improvements target legacy languages such as COBOL, Fortran and Visual Basic.
Why it matters: Higher accuracy in legacy code handling can streamline modernization of critical financial and enterprise systems.
Read the coverage
More in AI
AI10h ago4 sources
AI9h ago2 sources
