Jev-Powered Sushi Survivors Controller

Kimi K3 built a system that sends game state to Jev and converts Jev’s probability outputs into in-game actions. The system cleared the beginner BOSS stage of Sushi Survivors after 10 minutes.

View on X

KANA｜東京AI映像@KanaWorks_AI𝕏

Kimi K3 が、Jev に渡すゲーム状態の設計から、回答をゲーム操作へ変換するデコード処理まで一貫して構築。 Jev が返すのは確率値だけ。その出力をもとに、ゲーム内の行動を決定していきます。そして10分後—— 『寿司サバイバーズ』の初級BOSSステージをクリアしました🫡🤣 https://t.co/Hh1n6hZQBn

Sep 20, 2026X postsView on X

Kimi K3 handled the design of the game state passed to Jev and the decoding logic for turning Jev’s answers into actions. Jev returns probability values, which the system uses to decide what to do in the game.

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.