--- title: OpenAI’s COBOL migration score has made a striking leap description: GPT-6.1 Sol scored 86.6% accuracy on OpenAI's own COBOL migration test at low reasoning effort, up from GPT-5.6 Sol's 5.4%, while costing 39% less. At higher... canonical: https://daily.dev/posts/openai-s-cobol-migration-score-has-made-a-striking-leap-smwh0snvj twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: OpenAI’s COBOL migration score has made a striking leap | daily.dev og:description: GPT-6.1 Sol scored 86.6% accuracy on OpenAI's own COBOL migration test at low reasoning effort, up from GPT-5.6 Sol's 5.4%, while costing 39% less. At higher... og:url: https://daily.dev/posts/openai-s-cobol-migration-score-has-made-a-striking-leap-smwh0snvj og:image: https://api.daily.dev/og/posts/SmwH0SnvJ.png og:image:alt: OpenAI’s COBOL migration score has made a striking leap og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # OpenAI’s COBOL migration score has made a striking leap **[The New Stack](https://daily.dev/sources/newstack)** · 4 min read · 0 upvotes · 0 comments ## Summary GPT-6.1 Sol scored 86.6% accuracy on OpenAI's own COBOL migration test at low reasoning effort, up from GPT-5.6 Sol's 5.4%, while costing 39% less. At higher reasoning effort it reached 97.9%. OpenAI has not published the COBOL test cases, so developers can't independently verify the results. It did release MultiLangBench on GitHub, a separate test that has models build a SQLite-compatible database engine across 20 languages; GPT-6 Astra averaged 87.4% pass rate there, versus 26.8% for GPT-5.6 Sol. OpenAI itself notes production use still needs expert review against business rules. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What accuracy did GPT-6.1 Sol achieve on OpenAI's COBOL migration benchmark compared to GPT-5.6 Sol? GPT-6.1 Sol reached 86.6% accuracy at low reasoning effort on OpenAI's COBOL migration test, up sharply from GPT-5.6 Sol's 5.4%, while costing 39% less to run. At higher reasoning effort, GPT-6.1 Sol's accuracy climbed to 97.9%. The migrations target C++, Java, Kotlin, Python, and Rust and must pass hidden tests requiring byte-for-byte parity on observable outputs. _Teams planning COBOL modernization projects track model benchmark claims like this one on daily.dev._ ### Has OpenAI published the test cases for its COBOL migration benchmark? No, OpenAI has not released the underlying test cases or dataset for its COBOL migration evaluation, unlike its MultiLangBench suite which is public on GitHub. This means developers cannot independently inspect or reproduce the COBOL results, even though OpenAI credits feedback from developers and enterprise customers for guiding the work. _Developers weighing AI vendor benchmark claims can follow scrutiny like this on daily.dev._ ### How did GPT-6 Astra perform on OpenAI's MultiLangBench across 20 programming languages? GPT-6 Astra, OpenAI's flagship model, achieved an 87.4% average test pass rate across 20 programming languages on MultiLangBench, up from GPT-5.6 Sol's 26.8%. It scored above 83% on 19 of the 20 languages. The cheaper GPT-6 Luna model averaged only 40.3% but still beat GPT-5.6 Sol on every language at roughly 78% lower estimated API cost. _Developers comparing model cost and accuracy trade-offs across languages can track results like these on daily.dev._ ## Similar posts on daily.dev - [GPT-6 Astra Benchmarks: 99.9% or 62.7%? The Full ARC-AGI-3 Story](https://daily.dev/posts/gpt-6-astra-benchmarks-99-9-or-62-7-the-full-arc-agi-3-story-1dabqghdx) · Medium · 1 upvotes · 0 comments - [OpenAI launches GPT-6 Sol and Luna, boasting lower cost and fewer mistakes](https://daily.dev/posts/openai-launches-gpt-6-sol-and-luna-boasting-lower-cost-and-fewer-mistakes-7fzpvatba) · TechCrunch · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#openai](https://daily.dev/tags/openai) [View this post on daily.dev](https://daily.dev/posts/openai-s-cobol-migration-score-has-made-a-striking-leap-smwh0snvj) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"OpenAI’s COBOL migration score has made a striking leap","url":"https://daily.dev/posts/openai-s-cobol-migration-score-has-made-a-striking-leap-smwh0snvj","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/openai-s-cobol-migration-score-has-made-a-striking-leap-smwh0snvj"},"datePublished":"2026-10-22T15:42:17.381Z","dateModified":"2026-10-22T15:52:27.462Z","description":"GPT-6.1 Sol scored 86.6% accuracy on OpenAI's own COBOL migration test at low reasoning effort, up from GPT-5.6 Sol's 5.4%, while costing 39% less. At higher...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/a2388eb0f5440b6e96b567a594bcc3db?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/a2388eb0f5440b6e96b567a594bcc3db?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"The New Stack","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"The New Stack","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/newstack","url":"https://daily.dev/sources/newstack"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/openai-s-cobol-migration-score-has-made-a-striking-leap-smwh0snvj","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,openai","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"The New Stack","item":"https://daily.dev/sources/newstack"},{"@type":"ListItem","position":3,"name":"OpenAI’s COBOL migration score has made a striking leap"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/openai-s-cobol-migration-score-has-made-a-striking-leap-smwh0snvj#faq","mainEntity":[{"@type":"Question","name":"What accuracy did GPT-6.1 Sol achieve on OpenAI's COBOL migration benchmark compared to GPT-5.6 Sol?","acceptedAnswer":{"@type":"Answer","text":"GPT-6.1 Sol reached 86.6% accuracy at low reasoning effort on OpenAI's COBOL migration test, up sharply from GPT-5.6 Sol's 5.4%, while costing 39% less to run. At higher reasoning effort, GPT-6.1 Sol's accuracy climbed to 97.9%. The migrations target C++, Java, Kotlin, Python, and Rust and must pass hidden tests requiring byte-for-byte parity on observable outputs. Teams planning COBOL modernization projects track model benchmark claims like this one on daily.dev."}},{"@type":"Question","name":"Has OpenAI published the test cases for its COBOL migration benchmark?","acceptedAnswer":{"@type":"Answer","text":"No, OpenAI has not released the underlying test cases or dataset for its COBOL migration evaluation, unlike its MultiLangBench suite which is public on GitHub. This means developers cannot independently inspect or reproduce the COBOL results, even though OpenAI credits feedback from developers and enterprise customers for guiding the work. Developers weighing AI vendor benchmark claims can follow scrutiny like this on daily.dev."}},{"@type":"Question","name":"How did GPT-6 Astra perform on OpenAI's MultiLangBench across 20 programming languages?","acceptedAnswer":{"@type":"Answer","text":"GPT-6 Astra, OpenAI's flagship model, achieved an 87.4% average test pass rate across 20 programming languages on MultiLangBench, up from GPT-5.6 Sol's 26.8%. It scored above 83% on 19 of the 20 languages. The cheaper GPT-6 Luna model averaged only 40.3% but still beat GPT-5.6 Sol on every language at roughly 78% lower estimated API cost. Developers comparing model cost and accuracy trade-offs across languages can track results like these on daily.dev."}}]} ```