mizorewww/laya-mlx
Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.
From the publisher
README & documentation
Source previewRead the project’s overview, installation instructions and usage examples. The original README is the source of truth.
Open-weight typed decisions, running natively on Apple Silicon. 13.4 ms median end-to-end for a short English typed decision. 7.4 ms with the multilingual checkpoint. 0 output tokens. Local MLX inference, with no PyTorch, Transformers runtime, or cloud API. 中文 · Benchmarks · Snake demo · Hugging Face weights The GIF is an original-speed render of a real local Snake run. Every move calls Laya; the visible cycle safety layer can correct unsafe proposals. The latency figures above are the separate one-question API benchmark, not the frame time of the three-question Snake loop. Watch the 30-second MP4 · Snake speed and stability. Apple Silicon, Python 3.11+, macOS 14+. First load downloads the checkpoint; later inference is fully local. The measured environment is macOS 27.2, Python 3.12.13 and MLX 0.32.2. That MLX release supplies macOS 14, 15 and 26 wheels; the local installer selected the 26 wheel. Older supported macOS versions were…
Read the full README ↗ · Preview checked 2026-09-20T08:23:42.162Z
Inside the original README — Document outline
- Laya-MLX
- Quick start
- Performance on M3 Max
- Why typed decisions?
- Supported checkpoints
- Development install
- Python API
- Language routing and presets
- Command line
- Export an MLX checkpoint
- Tests and benchmarks
- Performance research
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the README
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: Apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
What this repository does
Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.
Repository facts
- Owner
- mizorewww
- Primary language
- Python
- Stars
- 2,059
- Forks
- 105
- Open issues + pull requests
- 4
- License
- Apache-2.0
- Archived
- No
- Default branch
- main
- Created
- 2026-09-19T13:50:26.000Z
- Last push
- 2026-09-19T16:32:30.000Z
Topics and intended use
Owner-supplied topics: apple-silicon, decision-model, inference, laya, local-ai, machine-learning, mlx, modernbert, system-one, typed-decisions
Review the README for scope, installation, examples and limitations. We do not run repository code or certify it.
Evaluate before installing
Review licensing and dependencies, inspect recent commits and unresolved issues, and test in an isolated environment before production use. Stars and forks alone cannot answer these questions.
Source and freshness
Source: GitHub. Metadata observed 2026-09-21T06:41:44.675Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.
Related repositories
EbookFoundation/free-programming-books
donnemartin/system-design-primer