{"version":1,"type":"story","url":"https://digestai.news/story/openai-announces-gpt-6-1-sol-with-claimed-86-6-to-97-9-cobol-migration","json":"https://digestai.news/story/openai-announces-gpt-6-1-sol-with-claimed-86-6-to-97-9-cobol-migration.json","markdown":"https://digestai.news/story/openai-announces-gpt-6-1-sol-with-claimed-86-6-to-97-9-cobol-migration.md","slug":"openai-announces-gpt-6-1-sol-with-claimed-86-6-to-97-9-cobol-migration","headline":"OpenAI announces GPT-6.1 Sol with claimed 86.6% to 97.9% COBOL migration accuracy","summary":"OpenAI introduced GPT-6.1 Sol on September 22, 2026 as the middle tier of its GPT‑6 family, available via the API and rolling out in ChatGPT Work and Codex for eligible Plus, Pro, Business, Enterprise and Edu plans. The company’s internal benchmark, reported by The New Stack on October 22, says the model achieved 86.6% accuracy on a low‑effort COBOL migration test and 97.9% when more reasoning effort was applied, a jump from the prior model’s roughly 5% score. OpenAI notes the numbers come from an undisclosed internal dataset and have not been independently reproduced, and it cautions that expert review is still required before production use.\n\nThe model offers a 1,050,000‑token context window and five reasoning‑effort settings that trade speed for accuracy. Pricing remains $2.00 per million input tokens, $10.00 per million output tokens, with cached input at $0.10 per million tokens. Compared with GPT‑6 Astra, OpenAI claims GPT‑6.1 Sol delivers near‑flagship performance at lower cost, while GPT‑6 Luna stays the low‑cost, high‑volume option.\n\nFor banks, insurers and government IT shops that still run legacy COBOL systems, the reported accuracy could make modernization projects more affordable, but the lack of external verification means pilots are advisable before large‑scale adoption.","keyPoints":["OpenAI’s internal test reports 86.6% low‑effort and 97.9% high‑effort COBOL migration accuracy for GPT‑6.1 Sol","GPT‑6.1 Sol provides a 1,050,000‑token context window and five reasoning‑effort levels","Pricing is $2 input/$10 output per million tokens, with cached input at $0.10 per million tokens"],"whyItMatters":"If the accuracy claims hold, GPT‑6.1 Sol could dramatically lower the cost and risk of modernizing legacy COBOL systems that still power banking, insurance and government services.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["OpenAI","Anthropic","The New Stack"],"models":["GPT-6.1 Sol","GPT-6 Sol","GPT-5.6 Sol","GPT-6 Astra","GPT-6 Luna","Claude Opus 5.5"],"people":[]},"firstPublishedAt":"2026-10-24T14:45:00Z","updatedAt":"2026-10-24T14:45:00Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"online-tech-tips.com","title":"OpenAI’s GPT-6.1 Sol shows dramatic improvements in legacy code migration including COBOL","url":"https://online-tech-tips.com/openai-gpt-6-1-sol-cobol-legacy-code-migration","publishedAt":"2026-10-24T14:45:00Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":{"title":"OpenAI GPT-6 Legacy Code Migration Saga","url":"https://digestai.news/thread/opinion-gpt-6-1-sol-may-be-the-best-default-model-for-codex-users","storyCount":6},"cite":{"text":"Digest AI, \"OpenAI announces GPT-6.1 Sol with claimed 86.6% to 97.9% COBOL migration accuracy\", 24 October 2026, https://digestai.news/story/openai-announces-gpt-6-1-sol-with-claimed-86-6-to-97-9-cobol-migration","publisher":"Digest AI","title":"OpenAI announces GPT-6.1 Sol with claimed 86.6% to 97.9% COBOL migration accuracy","datePublished":"2026-10-24T14:45:00Z","url":"https://digestai.news/story/openai-announces-gpt-6-1-sol-with-claimed-86-6-to-97-9-cobol-migration"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}