devwithjev
— reading now— views
Submit a build

Plays Games with Apple Foundation Models

The author hacked Apple Foundation Models to extract logits and use them to play a game without generating text.

お寿司@小説@osushi_cr𝕏
Jevって要は、Logitsさえ取れれば生成せずに判断できるんよね。ということでApple Foundation ModelsをハックしてLogitsを取り出し、ゲームをプレイさせてみた。 Macにはじめから入ってるLLMでこれだけのスピードが出るのすごくない? #AppleIntelligence #Jev https://t.co/1Qu397OGxE
Sep 20, 2026X postsView on X
The author says the model is built into the Mac and remarks on its speed. The post explains that logits can be used to make decisions without generation.

Also filed under Games & real time

  • Benchmarks Jev on Pokémon Red

    The project benchmarks Jev, TypeSafe’s fast decision model, against Jev paired with a GPT-6 Sol planner for playing Pokémon Red.

  • Plain-English Cellular Ecosystem Simulator

    LifePot is a cellular automaton-inspired ecosystem that users describe in plain English. Jev turns those inputs into species, feeding relationships, reproduction strategies, and environmental conditions.

  • Mapping Children's Behavioral Intent in Games

    Mores Research used Jev to map children’s behavioral intent in games and examine how it correlates with real-world behavior. It reports mapping intents across 45+ real sessions involving 20 child users over a long horizon.

  • StarCraft Agent for Fast Tactical Decisions

    Jev_Star is a StarCraft agent experiment that uses Jev for fast tactical decisions.