Reading the repo
Opening the box…
Reading the repo
Opening the box…
How to install on macOS
A nano replica of Jev: parallel decisions, dynamic candidates, and an end-to-end training pipeline.
Saved setup from 1 days ago
Will it work on macOS?
Heads-upIt also needs 1 setting in .env (1 secret).
Then run bash setup.sh. It stops at the first error and is safe to run again. Servers and tests aren't started; it lists them at the end. Read it first.
Needs Git: xcode-select --install
git clone https://github.com/TianyuCodings/NanoJev.git cd NanoJev
Node.js >=22, Python 3
# needs nvm: brew install nvm (or https://github.com/nvm-sh/nvm) nvm install --lts brew install python
Fill in secrets: AI_GATEWAY_API_KEY
cp .env.example .env
npm ci
git clone https://github.com/TianyuCodings/NanoJev.git
cd NanoJev
python -m pip install -r requirements-toy.txt huggingface_hub
Download the current checkpoint and data:
from huggingface_hub import snapshot_download
snapshot_download(
repo_id="C-Tianyu/NanoJev",
revision="unified-games-v1",
local_dir="checkpoints/NanoJev-unified",
allow_patterns=["best.safetensors", "config.json", "tokenizer/*", "backbone_config/*"],
)
snapshot_download(
repo_id="C-Tianyu/NanoJev-Data",
repo_type="dataset",
revision="unified-games-v1",
local_dir="data/NanoJev-unified",
)
Start inference in a CUDA environment:
python scripts/serve_decisions.py \
--checkpoint-dir checkpoints/NanoJev-unified \
--web-root web --port 8765 --disable-native-triton
The service loads the model once. Send state/question batches to POST http://127.0.0.1:8765/api/evaluate.
To explore the recorded games locally:
python3 -m http.server 8080 --bind 127.0.0.1 --directory web
Open http://127.0.0.1:8080/dev/?autoplay=1 for ViZDoom Basic or http://127.0.0.1:8080/dev/side-by-side.html?autoplay=1#maze for Maze.
No reports for macOS yet.
GitManual AI Agent reads the 3 key files and writes a step-by-step guide for macOS. Once written, it's saved for everyone.