Docs / source reviewed
Jev Go
Choose a move from legal candidates in a 9×9 Go game
How this project uses Jev
- Input
- A text board state, recent moves and legal candidates annotated with tactical features
- Jev decides
- White’s next intersection or pass, with a distribution over candidates
- Code executes
- Game code handles captures, legality, ko and scoring, samples from the returned distribution and falls back to a local heuristic on failure
Evidence and limitations
Missing credentials, network errors, low confidence or illegal choices trigger local AI. A complete game does not prove Jev played every turn. The model does not read board images; we did not run games or evaluate playing strength.
This project has not been run independently here. Author-reported results are not independently verified results.