OpenAI said on October 19 that its latest models improved at building and migrating software in programming languages with less code and documentation, including COBOL, Fortran, and Visual Basic.

It is also releasing MultiLangBench on GitHub, an evaluation that tests whether models can build a SQLite-compatible database engine across 20 programming languages.

OpenAI said GPT-6 Astra reached an average test pass rate of 87.4% at maximum reasoning effort, up from 26.8% for GPT-5.6 Sol.

In a separate COBOL migration evaluation, GPT-6.1 Sol achieved 86.6% accuracy at low reasoning effort, versus 5.4% for GPT-5.6 Sol at the same setting.

OpenAI said these languages have less code and documentation available for training than languages such as Python and JavaScript, but still run core systems.

It added that production migrations still require expert review and testing against business rules and system dependencies.

The update comes amid broader discussion of AI-assisted updates to legacy systems, including Anthropic’s Claude Code and its potential role in modernizing COBOL.

🔗 Source: OpenAI

Recent OpenAI developments