Bleepingcomputer OpenAI's GPT-6.1 Sol Enhances Legacy Code Migration Capabilities
Article Content
- •GPT-6.1 Sol shows 86.6% accuracy in migrating COBOL code to modern languages.
- •Significant improvements were noted across multiple legacy programming languages.
- •OpenAI's advancements could help modernize critical infrastructure reliant on outdated code.
OpenAI announced significant improvements in its GPT-6.1 Sol model for migrating legacy code, particularly in older programming languages like COBOL, Fortran, and Visual Basic. The model achieved an accuracy of 86.6% for migrating COBOL to modern languages, a substantial increase from 5.4% with the previous GPT-5.6 Sol. OpenAI's testing involved creating a SQLite-compatible database engine across 20 languages, where GPT-6 Astra scored an average pass rate of 87.4%. This advancement is crucial as COBOL underpins a large portion of online banking and ATM transactions. OpenAI emphasized that while these models show promise, expert review and testing are still necessary for production migrations. The improvements could aid companies in modernizing their software infrastructure, which is essential for maintaining critical services globally.
Ask AI about this cluster
Answers cite the sources they use
Timeline
More articles in this cluster (3)
Following this threat?
Track Anthropic in your own feed — you're alerted when they show up in new reporting, leak sites or exploitation.
Free account · no card needed