{"version":1,"type":"story","url":"https://digestai.news/story/typesafe-ai-s-jev-beats-pokemon-red-in-under-a-week","json":"https://digestai.news/story/typesafe-ai-s-jev-beats-pokemon-red-in-under-a-week.json","markdown":"https://digestai.news/story/typesafe-ai-s-jev-beats-pokemon-red-in-under-a-week.md","slug":"typesafe-ai-s-jev-beats-pokemon-red-in-under-a-week","headline":"TypeSafe ai's Jev beats pokemon Red in under a week","summary":"Jev, a decision model from TypeSafe AI, defeated Pokémon Red’s Elite Four and Champion, entering the Hall of Fame on September 23, 2026, after a week of play. The project was announced by Andrew Boyd, founder of Standard Agents Inc., and livestreamed in a browser or terminal with Jev moderating chat.\n\nJev selects from a list of options and returns solutions with a confidence figure, but it does not read the screen or produce text or images. Anthropic’s Claude Opus 5 monitored the game log and coached Jev by adjusting options and wording. Opus reduced the text sent to Jev by about two‑thirds and changed the request frequency to every six seconds. Human viewers also sent tips that were incorporated into Jev’s option list. The harness changelog recorded 474 entries, many failures such as walking into Lorelei’s entrance 53 times.\n\nOther runs followed. Christian Mathiesen at Frigade used Jev with a memory‑reading harness, estimating a cost of about $1–1.70 per 24 hours. A separate experiment by stmonty trained a small world model on an RTX 3080 Ti with more than 42,000 frames; it picked a starter in 52 of 100 tries. Jev launched on September 15, and LangChain described it as having an “outsized response.” Anthropic’s Claude Plays Pokémon stream, running Opus 4.5, had not finished Red as of January, but with Jev playing and Claude writing rules, the developer achieved victory.","keyPoints":["Jev, a decision model, beat Pokémon Red’s Elite Four and Champion in a week","Claude Opus 5 coached Jev by adjusting options and reducing text by two‑thirds","Christian Mathiesen’s Frigade run cost about $1–1.70 per 24 hours"],"whyItMatters":"The demo shows how specialized decision models, guided by an LLM coach, can solve complex games faster than general chatbots, highlighting the potential of hybrid AI systems for problem‑solving tasks.","category":{"slug":"agents","name":"Agents & Tools","url":"https://digestai.news/category/agents"},"entities":{"companies":["TypeSafe AI","Anthropic","Standard Agents Inc.","Frigade"],"models":["Jev","Claude Opus 5","Claude Opus 4.5"],"people":["Andrew Boyd","Christian Mathiesen","stmonty"]},"firstPublishedAt":"2026-09-27T11:30:00Z","updatedAt":"2026-09-27T11:30:00Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"Tom's Hardware","title":"Developer says Jev decision model beat Pokémon Red in under a week","url":"https://tomshardware.com/tech-industry/artificial-intelligence/developer-says-jev-decision-model-beat-pokemon-red-in-under-a-week-non-llm-engine-succeeds-where-traditional-chatbots-stalled-for-months-but-claude-opus-5-coached-the-model-through-its-dead-ends","publishedAt":"2026-09-27T11:30:00Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"TypeSafe ai's Jev beats pokemon Red in under a week\", 27 September 2026, https://digestai.news/story/typesafe-ai-s-jev-beats-pokemon-red-in-under-a-week","publisher":"Digest AI","title":"TypeSafe ai's Jev beats pokemon Red in under a week","datePublished":"2026-09-27T11:30:00Z","url":"https://digestai.news/story/typesafe-ai-s-jev-beats-pokemon-red-in-under-a-week"},"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"}