Skip to content

Jev Plays Pokemon — milanboers

Autonomous Pokemon Red demo where Jev reads a compact text state each turn and answers typed Choice and Noul questions that deterministic code maps to PyBoy button presses.

Stars
5
Language
Python
Published by
Community
Author
milanboers
Last verified
Oct 2026

About this project

A small autonomous Pokémon Red agent. It uses TypeSafe's System One model, Jev: Jev reads the game state as text, answers typed questions each turn, and deterministic code turns those answers into button presses on a PyBoy Game Boy emulator.

Quoted from the project's README. Read the full README.

Highlights

  • No vision or chat history; each turn is a fresh request built from RAM reads, tilemap dialog, and walkability maps.
  • Harness re-injects short-term memory with recent dialog hints and recent actions plus explored fraction.
  • High-level goals route through A* pathfinding with anti-pacing, stuck detection, and exit probing.
  • Today reliably completes the opening from bedroom to Oak lab starter pick and back to Pallet Town.

Watch out

Text-only reconstruction from RAM and tilemaps, not vision. Requires Python 3.14 or later, uv, your own legally obtained Pokemon Red ROM, and TYPESAFE_API_KEY; turbo mode removes the 60 fps cap so the per-decision API call becomes the pacing factor. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

Other listings of this project

Forks and copies of this project in the directory. This entry is the canonical one.

1GitHub stars
A Pokemon Red agent that boots a ROM, picks each move with Jev, and streams video plus a live decision panel from a small read-only local HTTP server.
Jev Demos & Experiments#community#jevbest#catalog-import

More Jev resources like this

49GitHub stars
StarCraft II research repo combining Jev action selection with optional Astra plans across full-game macro control and SMAC-Hard micromanagement, with videos, replays, and a paper.
Jev Demos & Experiments#community#jevbest#catalog-import
22GitHub stars
Terminal harness that checks ROS 2 robots with deterministic health tests, capability descriptions, SQLite history, and optional Jev evidence routing.
Jev Demos & Experiments#community#jevbest#catalog-import
19GitHub stars
MCP server where one browser_goal call hands a URL, a goal, and code-checked assertions to a server-side loop driven by a decision model over live page elements.
Jev Demos & Experiments#community#jevbest#catalog-import
16GitHub stars
Local Qwen3.5 decision models in 0.8B and 2B sizes with fitted temperatures, systemone-shaped server, playground, guard, MCP tools, and eval harness.
Jev Demos & Experiments#community#jevbest#catalog-import
22GitHub stars
Independent open-source local decision model inspired by Jev, with a spark-s1 family plus smaller control models for agent gates, served locally with probabilities and policy actions.
Jev Demos & Experiments#community#jevbest#catalog-import
114GitHub stars
Browser agent where Jev picks each action from indexed page candidates while an LLM plans and reads, returning cited answers or schema-checked data with credential controls.
Jev Demos & Experiments#community#jevbest#catalog-import

From the Jev guides

Original write-ups on what this entry is for.

A game tick is a state plus a legal action list. How Clash Royale, Minecraft, and the earlier Mario, Pokémon, StarCraft, and drone loops ask Jev one step at a time.
Jev Demos & Experiments#gaming#agents#control
Read article
Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai