GoKawiilStandard Agents' Jev decision model completes Pokémon Red in under a week
Andrew Boyd, founder of Standard Agents Inc., says his Jev decision model beat Pokémon Red's Elite Four and Champion, reaching the Hall of Fame on September 23, 2026, less than a week after the run began. Jev isn't a chatbot or LLM; it picks from lists of weighted options rather than reading the screen, while Anthropic's Claude Opus 5 monitored the game log and adjusted those options to help it when stuck. A harness changelog logged 474 fixes, including repeated failures like walking into a shut door 53 times and looping a ladder 124 times.
GoKawiil's interpretation of the reporting above, not reported fact.
The result highlights a different approach to AI game-playing than the chatbot-driven runs that have taken weeks or months to finish similar Pokémon titles, suggesting decision models paired with LLM oversight could solve certain tasks faster than pure language-model reasoning. However, since Claude Opus 5 actively reshaped Jev's options mid-game, the achievement reflects a hybrid human-assisted system rather than an autonomous model succeeding unaided, which could raise questions about how such benchmarks should be judged going forward.
- Jev, a decision model from Standard Agents Inc., finished Pokémon Red on September 23, 2026, in under a week.
- Claude Opus 5 acted as an indirect coach, editing the options Jev could choose from when it got stuck.
- A 474-entry changelog shows Jev repeatedly failing at tasks like doors and ladders before the harness was adjusted.
Nintendo Switch Online + Game Boy App — This story is all about AI conquering Pokémon Red, so why not relive the original classic yourself? With Nintendo Switch Online's Game Boy library, you can play the actual Kanto adventure that inspired the experiment, no coaching AI required. It's a nostalgic way to see just how tricky those Elite Four battles really are.
See Nintendo Switch Online + Game Boy App on Amazon → Affiliate link — we may earn a commission on purchases, at no extra cost to you. Product picked by AI based on this article; it is not a tested recommendation.Source: tomshardware.com — Shane Downing, 2026-09-27
Published there as: “Developer says AI decision model Jev beat Pokémon Red in under a week”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.