laya-mlx
Port of all three Laya checkpoints to Apple's MLX framework, for fast local System One decisions on Apple Silicon.
0 members list laya-mlx, unchanged over the last 12 weeks, data as of 7 November 2026
Members who list laya-mlx in their Stack, by week.
Members who list laya-mlx
Be the first to add laya-mlx
Nobody on Stackness runs laya-mlx yet. Add it to your stack and you're the first member running it.
About laya-mlx
Laya-MLX is a Python package that runs Laya typed decision models natively on Apple Silicon using the MLX framework. It enables developers to build systems that answer constrained questions, such as selecting from options, scoring against rubrics, or making true-false judgments, with calibrated probabilities. The tool generates no tokens and requires no text generation, PyTorch, or cloud API calls, making it suitable for fast local inference on Apple devices. The developer, mizorewww, distributes it under the Apache 2.0 license.
The package provides access to three pre-converted checkpoints: an English model, a multilingual model, and a typed-decisions workflow model. It handles different question types including choice, score, and boolean judgments, delivering answers with confidence metrics. Developers can describe options and rubric levels to improve decision quality. The tool runs fully locally after an initial checkpoint download and supports macOS with Apple Silicon and requires Python.
Checked 2 November 2026.
Nobody on Stackness runs laya-mlx yet.
Sign up and add itPosts about laya-mlx
- Llama.cpp alternatives for running local agents: ds4, Magnitude and Strata against llama.cpp, vLLM and Ollama
ds4 runs big MoE models on 96 GB Macs, Magnitude tunes kernels to your device, Strata puts one model on a gaming PC. Every speed claim is self-reported.
- Can a decision model prune your coding agent's context? What fast-jev-compaction keeps, costs and drops
fast-jev-compaction prunes in a second, not 30. At its default threshold it keeps almost no tool results, and a scorer that always says no cuts as much.
- How do you run System One decision models locally? Ollaya, laya-mlx and the runtime slot
To run a System One model locally, pick a runtime too: Ollaya serves ten open model families, laya-mlx runs Laya on Apple silicon, and llama.cpp needs a shim.