GoKawiilOpenAI reports major gains in legacy code migration with GPT-6.1 Sol and GPT-6 Astra
OpenAI says its newer models show large improvements writing and migrating code in older languages like COBOL, Fortran and Visual Basic. Using its new MultiLangBench test, GPT-6 Astra scored an average 87.4% pass rate versus 26.8% for GPT-5.6 Sol, with COBOL jumping from 10.9% to 72.1%. Separately, GPT-6.1 Sol reportedly reached 86.6% accuracy migrating COBOL to modern languages at low reasoning effort, rising to 97.9% at higher effort.
GoKawiil's interpretation of the reporting above, not reported fact.
These older languages still underpin large parts of banking, payments and scientific infrastructure, so better AI performance could ease long-standing modernization bottlenecks for institutions reliant on scarce COBOL and Fortran expertise. OpenAI's own benchmark and framing mean the results should be read as self-reported progress rather than independently verified capability. The company itself cautions that real-world migrations would still need expert human review, suggesting these tools are an aid rather than a full replacement for legacy system specialists.
- OpenAI's GPT-6 Astra scored 87.4% average on its new MultiLangBench test, up from 26.8% for GPT-5.6 Sol.
- COBOL performance rose from 10.9% to 72.1%, and Fortran from 21.4% to 83.5% between model generations.
- GPT-6.1 Sol reportedly reaches up to 97.9% accuracy migrating COBOL to modern languages, though OpenAI says expert review is still needed.
Source: bleepingcomputer.com, 2026-10-20
Published there as: “OpenAI says GPT-6.1 Sol is dramatically better at migrating legacy code”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.