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.
768 projects
gut-check
rfi-irfos
Uses Laya confidence to route uncertain agent-verification cases to a larger language model
crisis-triage
mazenDDr
Fine-tunes and evaluates Laya for multilingual disaster-message triage with uncertainty routing
vybir
Galactic717
Runs Laya locally with conformal error guarantees and includes score-based game-learning demos
laya
joldibaev
Demonstrates Laya classifying Russian customer requests and scoring urgency and churn risk
laya
vk-alto-none
Provides a multilingual non-autoregressive Laya decision engine with checkpoint routing
laya
lucasnpinheiro
Runs Laya in a Docker container and exposes predictions through an HTTP API and MCP server
LayaGUIDemo
SaturnAura
Provides a Gradio interface for asking Laya choice, score, and yes-or-no questions about text
LayaMCP
aydinozturk
Exposes Laya classification, triage, guardrail, and routing tools through an MCP server
layaSample
MatiasIac
Provides a local FastAPI playground for Laya decisions, including a chess interface
system1
cloudn1ne
Containerizes Laya as a self-hosted HTTP server compatible with the TypeSafe Jev API
LayaBasedAssistant
userasg
Uses Laya to route requests and gate risky actions in a local voice-and-text assistant
comfyui-laya
latent-inpainter
Adds custom Laya nodes to ComfyUI
laya-api
rykhalskyi
Provides an API server for Laya prediction requests with API-key management and rate limits
system-one-poc
tonysprite
Integrates Laya-based typed decisions into DeepSeek Harness tool approvals and selection
laya-service
42tr
Packages Laya's HTTP serving API in a Docker image
jevlaya
holiq
Integrates local Laya inference and hosted Jev into a provider-agnostic decision framework
pixel
adetbekov
Uses Laya to route robot-pet commands locally and Gemini for unknown requests
AudioLaya
ThanabordeeN
Trains a speech-to-decision prototype that classifies calls using Laya decision heads
LAYA-RLCD
shyamsridhar123
Teaches RLCD and trains Laya pilots through a tactical game, lessons, and benchmarks
laya-forge
devjothish
Fine-tunes and calibrates Laya decisions, then uses them to guard production agents
laya-test
JVMoreiraD
Evaluates the Laya model's decision-making capabilities
laya-mlx
Gates-456
Runs open-weight Laya typed decisions locally on Apple Silicon using MLX
laya-ultrafast
ShinyDataTech
Ports a browser-use agent to run Laya locally for browser action decisions
laya-windows
mateusgl8
Ports a terminal Snake game to Windows with real-time decisions from the Laya model
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