Laya for Python
The reference implementation of Laya is the `laya` package on PyPI, maintained by Convai Innovations under Apache 2.0. It ships three checkpoints, a Router that picks one per request by script and language, a CLI, an HTTP server and an optional MCP server.
1712 projects
hermes-nerve
keeltrace
Adds typed decision oversight and tool gating to Hermes agents using Jev or self-hosted Laya
laya-mps
afshinm
Runs Laya typed-decision inference locally on Apple Silicon using Metal Performance Shaders
laya-guard
scs0209
Locally judges AI coding-agent tool calls with regex checks and the Laya model
eikos
caiovicentino
Builds calibrated typed-decision models for finance and trading and evaluates them against Laya
JevAny
SimpleJev
Training and evaluation infrastructure for System 1 decision models, with Laya as a benchmark baseline
laya-onnx
Mattepiu
Exports a fine-tuned Laya decision model to ONNX for Python and Node.js inference
laya-multilingual-mlx
aac6fef
Converts the multilingual Laya decision model to MLX for Apple silicon
laya-fa-support
tarfandoon
Routes Persian customer-support messages to the appropriate department using a fine-tuned Laya model
llm-honing
HectorHHZ
Trains and evaluates small decision models through game-playing loops and shared decision-server integrations
edgejev
yzfly
Exports, quantizes, and serves Laya and other Jev-style models for offline CPU inference
ComfyUI-Laya
jtydhr88
Routes ComfyUI workflows using typed decisions from the Laya model
laya-mind2web-browser-agent
ShaunSpark
Fine-tunes Laya on Mind2Web for browser automation decisions
laya-multilingual-coreml-ane
aac6fef
Packages the multilingual Laya decision model as a Core ML bundle for Apple devices
MacJev-322M-4K-Laya
chaoliangUNSW
Fine-tunes Laya for long-context local Mac agents and compares it with the base model
laya-jev-lab
yibie
Compares Jev and Laya decision models and evaluates a local-first inference cascade
laya-browser-agent
ChenneyZhuang
Runs Laya locally to provide typed decisions for browser agents
laya-openvino
rupeshs
Adds an OpenVINO inference backend and serving support for Laya checkpoints
WechatVibe
tswawa
Analyzes WeChat conversations for intent, emotion, and participant profiles using Laya models
LayaStudio
biplovgautam
Fine-tunes Laya decision models locally on Mac and evaluates results before deployment
laya-vision-demo
thaitea
Demonstrates calibrated image-based yes/no, choice and rubric decisions with Laya Vision
opendecider
manjunathshiva
Provides calibrated System 1 decision models and benchmarks them against Laya and other models
laya-apple
tc3oliver
Runs Laya locally on Apple Silicon with MLX GPU and Apple Neural Engine support
laya-security-guardrail
shashvik
Runs local Laya typed decisions as cybersecurity inspection and guardrail checks
laya-coreml
aac6fef
Packages the Laya decision model as a Core ML bundle for Apple devices
Install
Python 3.10 or newer, per the core README:
python -m pip install laya
Optional extras: laya[serve] (HTTP server), laya[mcp] (MCP server), laya[langchain], laya[onnx] (ONNX Runtime) and laya[fast] (TileLang GPU fast path).
First decision
from laya import Router
router = Router() # downloads a checkpoint on first use
state = "Hi, we were billed twice for March. Please refund the duplicate today or we will cancel our plan."
questions = {
"department": {"type": "choice", "instructions": "Which department should handle this?",
"criteria": {"billing": "invoices, payments, refunds",
"technical": "bugs, outages, system errors",
"other": "everything else"}},
"urgency": {"type": "score", "instructions": "How urgent is this?",
"criteria": ["not urgent", "soon", "blocking"]},
"churn_risk": {"type": "noul", "instructions": "Does the user threaten to cancel or leave?"},
}
result = router.predict(state, questions)
print(result["answers"]["department"]["choice"]) # billing
print(result["routing"]["model"]) # english
The package also installs a laya command, for example laya "My payment failed twice" --preset triage.
Checkpoints
| checkpoint | encoder | params | context |
|---|---|---|---|
laya |
ModernBERT-large | 421M | 512 |
laya-multilingual |
mmBERT-base | 322M | 1024 (up to 8,192) |
laya-typed-decisions |
ModernBERT-large | 421M | 1024 |
The README reports 32.8 ms for one question with laya-multilingual on a Tesla T4, and 193 to 464 ms per request on CPU with checkpoints preloaded.
Other Python projects
laya-mlx and laya-coreml for Apple Silicon, laya-openvino for Intel CPUs, edgejev for INT8 ONNX on CPU, and laya-mcp for agent tools.
Caveats
The README is direct about limits: the base checkpoints score near chance on its typed-decisions benchmark (0.362 and 0.352), and the 0.766 figure comes from the fine-tuned checkpoint. Temperatures should be refit on your own data before you trust the probabilities. The English checkpoint collapses on non-Latin scripts, which is why the Router exists.
More ways to use Laya