[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-ai-agents-stumble-in-stardew-valley-benchmark":10,"sections":34},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":29,"feedback":33,"feedback_at":22,"cost_usd":33,"total_tokens":33},2809,"ai-agents-stumble-in-stardew-valley-benchmark","AI Agents Stumble in Stardew Valley Benchmark","A new benchmark called StarDojo puts AI agents inside Stardew Valley and finds even the best models succeed less than 13% of the time.","AI agents can beat world champions at chess and Go, but they cannot reliably tend a virtual farm.\n\nResearchers introduced StarDojo, a benchmark built on top of Stardew Valley, to test whether AI agents can handle the messy combination of practical tasks and social interaction that defines everyday life. The suite includes 1,000 curated tasks spread across farming, crafting, exploration, combat, and social relationships, plus a compact 100-task subset for faster evaluations. Agents get a unified interface that requires no keyboard or mouse emulation and can run multiple game instances in parallel. GPT-4.1, the top performer in testing, cleared just 12.7% of tasks - with failures concentrated in visual understanding, multimodal reasoning, and fine-grained control.\n\nThe result matters because most existing AI benchmarks treat cognition and social behavior as separate problems. StarDojo is one of the first to demand both at once, inside a single continuous environment, which is a closer approximation of how humans actually move through the world. A 12.7% ceiling from a state-of-the-art model suggests current multimodal agents are still far from the general competence their vendors market.\n\nStardew Valley is a gentle farming sim, not a hostile adversarial environment - which makes a sub-13% success rate from the best available model a useful reality check on how far the \"agentic AI\" narrative has outrun the underlying capability.","[\"ai\",\"benchmarks\",\"multimodal\",\"agents\"]","2026-06-30T04:00:00.000Z","2026-06-30T13:11:18.889Z","2026-06-30T13:11:21.858Z","published",null,[],"ai",[24,26,27,28],"benchmarks","multimodal","agents",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.07445",0,{"sections":35},[36,40,45,50,55,60,65,70,75,80,85,89,94,99],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",2590,"2026-07-16T04:00:00.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":44},"Security","security",294,"2026-07-15T19:59:48.000Z",{"name":46,"slug":47,"count":48,"latest_published_at":49},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":51,"slug":52,"count":53,"latest_published_at":54},"Policy","policy",158,"2026-07-16T00:02:48.000Z",{"name":56,"slug":57,"count":58,"latest_published_at":59},"Hardware","hardware",122,"2026-07-14T19:46:26.000Z",{"name":61,"slug":62,"count":63,"latest_published_at":64},"Consumer Tech","consumer-tech",93,"2026-07-13T13:20:48.000Z",{"name":66,"slug":67,"count":68,"latest_published_at":69},"Software","software",70,"2026-07-13T19:52:25.000Z",{"name":71,"slug":72,"count":73,"latest_published_at":74},"Science","science",66,"2026-07-10T10:29:37.000Z",{"name":76,"slug":77,"count":78,"latest_published_at":79},"Dev Tools","dev-tools",59,"2026-07-07T04:00:00.000Z",{"name":81,"slug":82,"count":83,"latest_published_at":84},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":86,"slug":87,"count":83,"latest_published_at":88},"Startups","startups","2026-06-29T20:55:50.000Z",{"name":90,"slug":91,"count":92,"latest_published_at":93},"General","general",29,"2026-07-10T22:28:58.000Z",{"name":95,"slug":96,"count":97,"latest_published_at":98},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":100,"slug":101,"count":102,"latest_published_at":103},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]