replica source-backed
PlayJev: a 0.8B visual decision model for ten games
An independent Jev-like model fine-tuned from Qwen3.5-0.8B-Base to choose legal actions from raw game frames. The project publishes model weights, training records and a gallery of ten browser games.
Notes
The author reports one forward pass per action, 43 ms per move on an H200, and training on 2.2 million frames; the model does not generate an action as text.
source-backed — Public repository, docs or live artifact. About this label
Destinations
How to run
- Stack
- HTML / JavaScript demo; Python model
- Requirements
- A modern browser for recorded playback
- Keys / accounts
- No account or API key for the public recorded demo.
Open the published demo and select a game tile to inspect a recorded model episode. Local model serving is a separate Python/CUDA setup documented in the README. Playback does not run live inference.
README review · · Documentation reviewed; project installation and live API not run.
Next: follow the practical guide → · compare projects for this task →
Verification levels
- HTTP availability
- Link reachable ·
Method: bounded HTTP request with redirects. HTTP 200 · HTTP response only. Checked destination ↗ · Last reachable - Browser demo scenario
- passed ·
Method: browser-automation. Scenario: Load the recorded Snake episode, pause playback, restart it and observe advancing steps and move probabilities. — Saved episode playback; live model inference was not tested. - Build
- Not checked
- Live API test
- Not checked