Laya on MLX
laya-mlx is an independent MLX port of Laya that runs the original weights on the Apple Silicon GPU with no PyTorch or Transformers runtime. It keeps upstream prompt formatting, calibration and output schema, and it is the base for many Mac demos, agents and benchmarks.
260 projects
obsidian-laya-tagger
tyPhoon-collab
Automatically tags Obsidian notes using the local Laya MLX typed-decision model
laya-mcp-server
ocha-no-taiko
Wraps the laya-mlx typed-decision model in an MCP server for tools such as Claude Code
laya-flappy-bird
GadhiyaRaj
A Flappy Bird-style game uses Laya to choose real-time flap or coast actions
laya-claude-code
harshadptl
Claude Code plugin that uses laya-mlx to pick the model tier and time context compaction
laya-or-jev-pong
bananadonn
Compares local Laya and hosted Jev decision APIs by having them control paddles in Pong
laya-mlx-jev-compat
matt-starburst
Serves a local MLX Laya model through a Jev-compatible HTTP endpoint
e2e-ui-test-laya
spencerlepine
A local browser-use agent uses Laya through MLX to choose actions and verify task completion
system-one
arhamj
Serves Laya typed decisions from Apple silicon over a private Tailscale endpoint
ai-update-radar-lab
starhunt
Tests Laya-based project relevance decisions and compares results with saved Jev evaluations
Jev
Caho1
A research repository documents Laya fine-tuning, evaluation, local inference and browser automation experiments
Tuantuan
JunbiaoXue
Provides a native macOS desktop companion that uses local Laya decisions to choose behavior
jev-like-model-playground
LiuWei997
A notebook template runs lightweight System 1 models, including Laya, locally on Apple Silicon with MLX
LayaBasedAssistant
userasg
Uses Laya to route requests and gate risky actions in a local voice-and-text assistant
Laya.jl
AtelierArith
Runs Laya typed-decision models in Julia on CPU or Apple GPU
laya-mlx
Gates-456
Runs open-weight Laya typed decisions locally on Apple Silicon using MLX
laya-doom
dylanbstorey
Runs Laya in a real-time control loop to play Doom on Apple Silicon
laya-js
johnhenry
Runs Laya typed decisions in JavaScript across MLX, WebGPU and CPU backends
laya-mlx-mcp
Blackphi6
Exposes Laya-MLX typed decisions through an MCP server for Apple Silicon
codex-laya-mcp
AetherZZ2
Exposes local Laya typed-decision tools to Codex through MCP on Apple Silicon
jev-laya-explore
NachiketKandari
Evaluates the local Laya decision model alongside TypeSafe's hosted Jev
laya-mlx-benchmarks
jayluxferro
Benchmarks Laya typed-decision models running through the MLX inference port
cam-laya-mcp
vdqvinh2004
Adds optional local Laya-MLX decision tools for coding agents
laya-snake-lab
shaunthebuilder
Combines a Snake game with Laya-powered autoplay, training, and live decision probabilities
laya-r-mlx
tjameswilliams
Provides a native Rust inference engine for Laya with MLX and candle backends
Install
Requirements from the laya-mlx README: Apple Silicon, Python 3.11+, macOS 14+.
pip install laya-mlx
Pre-converted FP16 checkpoints are on Hugging Face: aac6fef/laya-mlx, aac6fef/laya-multilingual-mlx and aac6fef/laya-typed-decisions-mlx. You can also load the original convaiinnovations/laya IDs.
Router and CLI
The port includes the upstream language router and presets:
from laya_mlx import Router, triage_questions
router = Router(dtype="float16", max_loaded=2)
result = router.predict({"message": "发票被重复扣款，请退款。"}, triage_questions())
print(result["routing"]) # multilingual
For the terminal demo, where Laya plays Snake:
pip install 'laya-mlx[demo]'
hf download aac6fef/laya-multilingual-mlx
laya-snake
Reported numbers
On an M3 Max, FP16, one short question: 13.42 ms P50 for the 421M English model and 7.39 ms for the 322M multilingual model. The README states that all three checkpoints matched the upstream answer on 63/63 validation questions in FP32 and FP16. That measures port fidelity, not task accuracy.
Projects built on laya-mlx
- laya-ultrafast: a port of browser-use/jev-ultrafast that makes browser decisions locally.
- laya-mcp (wsargent): an MCP server with triage, guard, moderation and email tools.
- laya-fast: adds Neural Engine execution next to the MLX GPU.
- laya-jev-lab and jev-laya-benchmark: independent Jev comparisons that use laya-mlx as the Laya side.
- laya-vs-jev: Laya and Jev playing Chrome's T-Rex game side by side.
Caveats
This is not an official Convai Innovations release. Training and fine-tuning stay in the upstream project. BF16 can be requested but is not part of the published validation.
More ways to use Laya