{"version":1,"type":"story","url":"https://digestai.news/story/openai-releases-gpt-6-sol-with-7-7-improvement-and-half-api-cost","json":"https://digestai.news/story/openai-releases-gpt-6-sol-with-7-7-improvement-and-half-api-cost.json","markdown":"https://digestai.news/story/openai-releases-gpt-6-sol-with-7-7-improvement-and-half-api-cost.md","slug":"openai-releases-gpt-6-sol-with-7-7-improvement-and-half-api-cost","headline":"OpenAI releases GPT-6 Sol with 7.7% improvement and half API cost","summary":"OpenAI released GPT-6 Sol (Max) on September 22, 2026, reporting a +7.7% net improvement across more than 4,000 real‑world agentic sessions on the Agent Arena leaderboard. The new model sits at #6, up from #8 for GPT‑5.6 Sol (xHigh). API pricing drops to $2 per million input tokens and $10 per million output tokens, a 50% reduction from GPT‑5.6 rates.\n\nOn the Confirmed Success metric GPT‑6 Sol scored +11.4% and ranked #4 overall, showing stronger reliability for completed tasks. On DeepSWE v1.1 it posted 68.8%, trailing Anthropic’s Claude Fable 5 but at about 80% lower cost. OpenAI also launched GPT‑6 Luna the same day, targeting scalable, cost‑effective agentic workflows, automation pipelines and always‑on AI infrastructure.\n\nThe pricing war continues with Anthropic’s Claude Opus 5 and Claude Fable 5 maintaining higher DeepSWE scores, while AutomationBench results suggest GPT‑6 Sol completes tasks at lower cost than some Claude Opus 5 configurations.","keyPoints":["GPT‑6 Sol achieves +7.7% net improvement on Agent Arena, up from GPT‑5.6 Sol","API pricing halves to $2/m input and $10/m output, a 50% reduction","DeepSWE v1.1 score 68.8%, 80% cheaper than top Claude models"],"whyItMatters":"The model’s cost‑performance gains could lower barriers for large‑scale agentic workflows, shifting the competitive balance in AI‑driven automation and prompting pricing wars among major vendors.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["OpenAI","Anthropic"],"models":["GPT-6 Sol (Max)","GPT-5.6 Sol (xHigh)","GPT-6 Luna","GPT-6 Astra","Claude Fable 5","Claude Opus 5"],"people":[]},"firstPublishedAt":"2026-09-25T14:11:00Z","updatedAt":"2026-09-25T21:36:00Z","sourceCount":3,"hasPrimarySource":false,"sources":[{"outlet":"cryptobriefing.com","title":"OpenAI’s GPT-6 Sol (Max) achieves 7.7% net improvement while cutting API costs in half","url":"https://cryptobriefing.com/openai-gpt6-sol-max-performance-cost","publishedAt":"2026-09-25T14:11:00Z","type":"press","primary":false,"lead":true},{"outlet":"note.com","title":"I actually tried using GPT-6 Sol/Luna in Codex","url":"https://note.com/eng_ai_workbook/n/na693bd3fcd3c?hl=en","publishedAt":"2026-09-25T21:36:00Z","type":"press","primary":false,"lead":false},{"outlet":"note.com","title":"Is AI Competing on 'Affordability' Rather Than 'Intelligence'? The Price War Started by GPT-6 and Claude on the Same Day","url":"https://note.com/funakkie/n/n4252df655fb3?hl=en","publishedAt":"2026-09-25T19:59:00Z","type":"press","primary":false,"lead":false}],"sourceNotes":null,"discussions":[],"thread":{"title":"OpenAI Unveils GPT-6 Sol and Luna Models","url":"https://digestai.news/thread/comparing-openai-models-on-aws-for-workload-efficiency","storyCount":4},"cite":{"text":"Digest AI, \"OpenAI releases GPT-6 Sol with 7.7% improvement and half API cost\", 25 September 2026, https://digestai.news/story/openai-releases-gpt-6-sol-with-7-7-improvement-and-half-api-cost","publisher":"Digest AI","title":"OpenAI releases GPT-6 Sol with 7.7% improvement and half API cost","datePublished":"2026-09-25T14:11:00Z","url":"https://digestai.news/story/openai-releases-gpt-6-sol-with-7-7-improvement-and-half-api-cost"},"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"}