LA
LAYA$LAYA
Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.
ShippingOn curve on Pons
Activity, last 13 weeks
Ship log, last 90 days
October 2026
- 10laya-mlx v0.4.0GitHub release in mizorewww/laya-mlx
- 104 code changes: Materialize permutation test parameters before inferenceGitHub commits in mizorewww/laya-mlx
- 2laya-mlx v0.3.0GitHub release in mizorewww/laya-mlx
- 21 code change: Sync upstream runtime fixes and automate v0.3.0 releasesGitHub commits in mizorewww/laya-mlx
September 2026
- 221 code change: Sync with upstream v0.3.5: temperature clamping, language routing fix…GitHub commits in mizorewww/laya-mlx
- 195 code changes: Launch Snake demo, opt-in runtime optimizations, and PyPI packagingGitHub commits in mizorewww/laya-mlx
Sources read
- mizorewww/laya-mlx11 of 12 recent commits counted as real work
This repository was linked by the token’s own on-chain metadata. It is not verified until the builder submits it. Merges, bot commits, dependency bumps, typo fixes and docs-only commits are not counted.
Badge for your README
[](https://shiplogpons.xyz/project/laya-97196a)The badge updates with the project’s status and momentum.