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.
770 projects
mcp-laya
dockndevai
Provides an MCP server exposing Laya typed-decision tools to agents
laya-server
roilanrodriguez55
Runs the real Laya server locally or in Docker and provides Bruno API tests
Laya-Finetune
zamax14
Fine-tunes and evaluates multilingual Laya for support-ticket classification
laya-plus
KotalaKishanReddy
Extends Laya's effective context with longer inputs and retrieve-then-decide processing
laya-rex
luiz0ar
A browser dashboard uses Laya decisions to control the T-Rex Runner game
laya-games
BM-Popcorn
Demonstrates Laya-driven turn-based games and fine-tunes the model on engine-labeled positions
laya-top
bsisduck
Monitors local processes with explainable tags and optional Laya inference
laya-universal
abusuraihsakhri
Provides cross-platform inference utilities for Laya checkpoints using ONNX Runtime and MLX
laya-finetuning
sothi-em
Fine-tunes Laya for key-value match detection using the KVP-10K dataset
laya-ecosystem
MiguelFAraujo
Provides utilities for supervision, context compaction, search, and UI decisions using Laya and Jev
laya_bio
maris205
Adapts the Laya encoder into a shared candidate-scoring model for biological sequence tasks
laya-server
zachary-walters
Serves Laya classification, scoring, and detection endpoints with Apple Silicon acceleration
hey-laya
touhidsiddiqueeraj-bit
Runs an offline voice assistant that uses Laya for on-device decisions
zabbix-laya
j3udiel
A local web lab evaluates fictional Zabbix alerts with Laya and lets users download test results
laya-studio
jiaheng6
Provides a Windows desktop client for installing, running, testing, and serving Laya models
laya_demo
itsvrushabh
Demonstrates and benchmarks Laya’s typed decision engine across interactive use cases
laya-deploy
iintothewind
Deploys and benchmarks Laya Serve on Windows with Docker and a temporary public tunnel
laya-studio
felix-homelab
Wraps Laya in a web panel for schema authoring, evaluation, automation, and calibration monitoring
Laya_wrapper
ugelucero
Wraps Laya checkpoints in a command-line tool and FastAPI service for request routing
everything-laya
its-negy-lmsy
Provides an MCP server and Python toolkit for self-hosted Laya
local-laya
maurostorch
Runs Laya as a Dockerized HTTP service with a web interface
laya-lab
Parswanadh
Coordinates long-context decision-engine research for the open Laya model
laya-indic-bench
jay123anta
Evaluates Laya decision models on Hindi and Assamese
laya-call-router
riefer02
Demonstrates fine-tuned Laya decisions in an inspectable dealership call-routing workflow
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