The OpenAI enhanced AI model conquered the popular programming language, and the Cobol migration accuracy soared to 97.9%.OpenAI 增强 AI 模型攻克冷门编程语言，COBOL 迁移准确率飙至 97.9%

IT House October 21st news, OpenAI released a research report yesterday (October 20th), saying that it enhances the ability of models such as GPT-6 Astra to build and migrate COBOL, Fortran, Visual Basic and other "low-resource programming languages". These languages, with less open code and documentation, support critical systems such as banking, ATMs, weather forecasting, and nuclear safety analysis. To measure competence, OpenAI launched two assessments: MultiLangBench required the model to build a SQLite-compatible database engine from scratch in 20 languages; another required the model to migrate COBOL programs to C + +, Java, Python, Rust and other languages while keeping the original business behavior unchanged. The results showed that the average test pass rate of GPT-6 Astra in 20 languages of MultiLangBench was 87.4%, a significant increase from 26.8% of GPT-5.6 Sol; COBOL increased from 10.9% to 72.1%, Fortran increased from 21.4% to 83.5%, and Visual Basic increased from 27.5% to 89.4%. The lower cost GPT-6 Luna has an average pass rate of 40.3%, and the estimated API cost is about 78% lower than GPT-5.6 Sol. The relevant picture is attached as follows: In the COBOL migration evaluation, the accuracy rate of GPT-6.1 Sol reached 86.6% at low inference strength, which is much higher than the 5.4% of GPT-5.6 Sol, and the cost is reduced by 39%; At higher inference strength, the accuracy rate reached 97.9%, close to the upper limit of the evaluation. OpenAI says this progress helps businesses maintain legacy systems, fix defects, and modernize software. However, the company also emphasized that the use of AI migration results for production environments still requires expert review and testing in conjunction with specific business rules and dependencies.