News
OpenAI says it is improving ChatGPT for low-resource programming languages, adding evaluations for software engineering, SQLite and COBOL migration.
OpenAI said it is improving ChatGPT across low-resource programming languages, adding new evaluations for software engineering, SQLite and COBOL migration, according to its Oct. 19, 2026 post titled Advancing Coding in Low-Resource Programming Languages. The update points to a narrower focus on coding performance and migration work rather than a broad announcement of new executable language support inside ChatGPT.
Evidence and context
The source material identifies three areas of evaluation: software engineering tasks, SQLite, and COBOL migration. OpenAI did not provide performance scores, benchmark results or a list of specific languages in the supplied text, so the announcement is best read as a tooling and assessment update rather than a quantified claim about model accuracy.
That distinction matters because the surrounding release history describes improvements to coding workflows and file handling, not runtime execution for arbitrary languages. In background material verified as of Oct. 19, 2026, OpenAI had previously expanded support for text and code file formats on Feb. 13, 2026, and added inline editing, previews of diagrams and mini apps, and split-screen code review on Feb. 19, 2026. Those releases improved how users interact with code in ChatGPT, but they did not establish that any programming language can run natively inside the product.
The October update also sits alongside other recent coding-related changes. OpenAI updated GPT-6 Pro in ChatGPT on Oct. 16, 2026 with faster responses, improved plugin and tool use, and parallel agents, while Codex received composer predictions beta on Oct. 9 and Codex Cloud on Sept. 29, 2026. The supplied context says these are separate coding workflows: ChatGPT is the conversational interface, while Codex is the repository-based development environment. The material also notes that OpenAI’s Oct. 7, 2026 Chat-tab GPT-6 rollout did not change the models powering Work and Codex.
No pricing, market data or user adoption figures were supplied with the announcement. The timing of any further language-specific rollout or benchmark disclosure is unknown from the provided material.
Credit: Source link
