OpenAI has released results for MultiLangBench, which asks models to build a SQLite-compatible engine in 20 languages, and ...
Interesting Engineering on MSN
OpenAI models make huge gains in modernizing decades-old software with 97.9% accuracy
OpenAI has significantly improved its models’ ability to write software in older programming languages ...
OpenAI announced significant performance improvements in its models for programming languages with relatively little code and ...
On October 19, 2026, OpenAI's official website announced an initiative to improve coding capabilities in "programming ...
OpenAI reports gains in both a 20-language coding benchmark and tests of COBOL migration, while noting that production ...
On October 19, 2026, an announcement posted by OpenAI on X caught my eye. It stated that the latest AI models are now capable ...
Redditユーザーが、スマートフォン・ブラウザ・サーバーが依存している23の主要なオープンソースプロジェクトの履歴から、プロジェクトのメンテナーについて記録する調査を行いました。その結果、23のプロジェクトのうち11では、過去1年間に定期的に変更を ...
SQLiteの実装評価で20言語平均87.4%を記録 ...
米OpenAIは10月19日、COBOLやFortran、 Visual Basic など学習データの少ない言語でのコーディング性能向上を発表した。新ベンチマーク「MultiLangBench」で GPT-6 Astra のCOBOL合格率は10.9 ...
OpenAI verilerine göre GPT-6 Luna, MultiLangBench testinde yüzde 40,3 başarı sağlarken GPT-5.6 Sol modeline kıyasla yüzde 78 ...
GPT-6 Astra的平均测试通过率为87.4%，而GPT-5.6 Sol为26.8%。在COBOL测试中，Astra的得分从10.9%升至72.1%；在Fortran测试中，则从21.4%升至83.5%。 所有受测语言的成绩均有所提升，包括COBOL（从10.9%升至72.1%）、OCaml（从25.6%升至85.8%）、Fortran（从21.4%升至83.5%）和Visual ...
IT时代网10月20日消息，OpenAI 近日发布文章，介绍其在低资源编程语言编码能力上的进展，并开源了名为 MultiLangBench 的多语言编程评测框架。所谓低资源语言，是指可用代码与文档远少于 Python、JavaScript 的编程语言 ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results