{"rewrite":{"id":"r_1796f332a95bc10efdcdd21b","clusterId":"c_47eece2df4ab134d4dcd868f","slug":"typesafe-ai-launches-jev-a-non-llm-model-with-zero-tool-call-errors","model":"deepseek-v4-1-flash","headline":"TypeSafe AI Launches Jev, A Non-LLM Model With Zero Tool-Call Errors","summary":"TypeSafe AI announced System One Models on September 15, 2026, along with Jev, the first model in the lineup. Jev skips sequential text generation and outputs a deliverable at once from a predefined structure. The company lists input at $0.042 per million tokens, output at $0, and says Jev matches GPT-5.6 Terra accuracy at about one-hundredth the cost.","whyItMatters":"A co-developer of ChatGPT is now selling an architecture that abandons the left-to-right generation step those systems depend on, and TypeSafe AI is putting the claim on a price sheet rather than a benchmark chart.","webCardHtml":"\u003cp\u003eTypeSafe AI is led by CEO Diogo Almeida, who was involved in ChatGPT\u0026#39;s development. The company says Jev outputs results all at once according to a predefined structure instead of continuing text, and that no errors occur in its outputs. Every output carries a confidence score, and tool-call error rate is given as 0%. An early access program is open, with documentation posted for System One Models.\u003c/p\u003e","blueskyPost":"Jev's output price is $0 per million tokens. At one-hundredth of GPT-5.6 Terra's cost, the model turns inference into a fixed input charge, which shifts where the spending pressure lands.","twitterPost":"TypeSafe AI prices Jev output at $0 and input at $0.042 per million tokens, so cost scales with what goes in, not what comes out.","threadsPost":"TypeSafe AI charges $0 for Jev output and $0.042 per million input tokens, while Jev matches GPT-5.6 Terra accuracy at about one-hundredth the cost. A zero output price makes the input side the only meter running, which is the part of the pricing table worth watching.","newsletterBlurb":"TypeSafe AI announced System One Models and Jev, a model that does not generate text sequentially. The company says it matches GPT-5.6 Terra accuracy at roughly one-hundredth the cost, with $0.042 per million input tokens and output listed as $0. Early access is open.","attributionJson":"[{\"source\":\"GIGAZINE\",\"url\":\"https://gigazine.net/news/20260916-system-one-jev/\",\"title\":\"ChatGPT Co-Developer Develops \\\"Jev,\\\" an AI Model That Processes Differently From LLMs, Matching GPT-5.6 Terra Performance While Running Tasks Cheaply and at Ultra-High Speed\"}]","lintFlagsJson":null,"lintHits":0,"costUsd":0,"inputTokens":4757,"outputTokens":580,"status":"published","repairAttempts":0,"nextRepairAt":null,"factsAttemptedAt":1789529367,"createdAt":"2026-09-16T03:25:07.000Z","publishedAt":"2026-09-16T03:28:22.000Z","updatedAt":"2026-09-16T03:28:22.000Z"},"cluster":{"id":"c_47eece2df4ab134d4dcd868f","canonicalTitle":"ChatGPT共同開発者がLLMとは異なる方法で処理するAIモデル「Jev」を開発、GPT-5.6 Terra級の性能でタスクを安価かつ超高速に実行可能","representativeArticleId":"a_9a21f2a271d517e196768f4d","sourceCount":1,"writtenSourceCount":1,"writeAttempts":0,"isSolo":true,"entitiesJson":"{\"anime_titles\":[],\"manga_titles\":[],\"work_titles\":[],\"studios\":[],\"people\":[],\"type\":\"news\",\"domain\":\"other\",\"is_roundup\":false}","contentType":"news","status":"published","firstSeenAt":"2026-09-16T02:42:00.000Z","lastSeenAt":"2026-09-16T02:42:00.000Z","updatedAt":"2026-09-16T03:28:20.000Z"},"attribution":[{"source":"GIGAZINE","url":"https://gigazine.net/news/20260916-system-one-jev/","title":"ChatGPT共同開発者がLLMとは異なる方法で処理するAIモデル「Jev」を開発、GPT-5.6 Terra級の性能でタスクを安価かつ超高速に実行可能"}],"entities":{"anime_titles":[],"manga_titles":[],"work_titles":[],"studios":[],"people":[],"type":"news","domain":"other","is_roundup":false},"keyFacts":null}