Game AI with decision models

Independent field guide to Jev 1.13.0 behaviour across 11,621 requests: framing sensitivity, Snake and 3D city driving experiments.

Factorio agent on FLE where Jev picks goals, next actions and stuck detection while code owns rules and actuation.

Nine-board gomoku workbench comparing how two Jev players respond to five different input representations of the board.

Craftax survival agent where Jev picks one Choice per step from code-listed actions, with optional LLM goals, plus ablations.

Snake research environment where game rules stay in code and Jev picks the movement direction from structured state each tick.

StarCraft II research repo where Jev Choices pick macro actions and unit micromanagement, optionally constrained by a larger model's plan.

Tetris where each reachable placement is an English sentence that Jev ranks, with controls showing a regex baseline does better.

RPG experiment pitting Jev against OpenAI's Decisions API as NPCs, comparing wins, per-move accuracy, calibration and cost.

Evaluates Jev as a fixed policy in Gymnasium and Atari games, picking a legal action each step via one Choice question.

Mineflayer Minecraft bot where code lists feasible actions and Jev picks one in about 0.2 s; 15 of 40 runs reached the Nether.

An LLM writes a task-specific Jev harness, improved with rewards and traces; a Pokemon harness rose from 25% to 75% wins.

Embodied Minecraft agent where a typed decision plane (Jev or Liquid d1) picks grounded transitions and a kernel owns verification.

Model Kombat: a 12,000-neuron fly connectome fights Jev in mk.js, with Jev picking one of seven moves about five times a second.

Pandu: tiny local policy networks (2.9K-5K parameters) with calibrated uncertainty that fall back to a teacher when unsure.

Chinese Slay the Spire 2 agent research: code handles rules, cheap Jev makes most choices, a stronger model takes key moments.

Tic-tac-toe teaching project comparing Jev Choice, Score and Noul designs across three levels of state detail.

Short evaluation finding that Jev plays chess poorly, often calling unbalanced positions draws.

Local app testing Jev as a chess decision-maker against Stockfish, with Stockfish grading every pick.

Gym-style environment for offline Warcraft III with deterministic stepping and fog-filtered observations, plus an agent using Jev for micro.

Repository describing a Roblox Studio instance reviewer using typed Jev criteria with citations; README is mostly marketing.

npm library for chess with Jev: natural-language move intents to legal moves, parallel move evaluation and historical persona opponents.

Dependency-free TypeScript library that converts raw app and game state into labelled bands and pruned lists that Jev reads reliably.

Penalty shootout game against Jev, run as a Render Workflows task with scores in Postgres, in TypeScript and Python versions.

RuneBench fork where Jev plays RuneScape by choosing from about 50 bounded rs-sdk actions and per-tick interventions.

Next.js community playground extending the 110-example TypeSafe library with A/B comparisons, routing, extraction, games and simulations.

Community web playground with 110 runnable Jev examples in 22 categories, including A/B comparisons, games and dilemmas.

Pokémon Red on PyBoy where code owns route and arithmetic, Jev picks only at branches, and battle faint predictions are Brier-scored.

Local Snake battle where four NanoJev AI snakes pick moves from a candidate distribution while you play the fifth.

Pygame chess where python-chess owns the rules and Jev picks one labelled legal move per turn, with Human or Jev on either side.

Playful web app where twelve absurd characters vote on your idea via one Jev request with 36 independent questions.