Routing with Laya
A router runs before every request, so its own latency and cost land on the critical path. Laya answers a routing question in one forward pass with no generated tokens, which makes it a natural fit for choosing an LLM tier, a support queue or a tool.
342 projects
jev-decision-studio
Hetgandhi25
Runs local Laya decision models for incident triage with an Ollama fallback
laya-triage
CodeCave0000
Adds local Laya-based message triage to the OpenClaw agent plugin
Laya-Workflow-Retrieval
sanjanabhat846
Uses Laya to select verification checklists for engineering code reviews
hermes-laya-tools
andrei-tolstov
Routes Hermes Agent tool calls and categorizes tools using Laya decision inference
llm-model-router
dmoonat
Routes Google ADK requests between models using Laya typed decisions
trustreflex
Maverick7
A calibrated local assistant learns from Ollama or Laya and routes trusted answers to a fast student
dsh-laya-mcp-skill-router
royalallen
Routes tools and agent skills using Laya decisions and local embedding-based recall
Sparrow-Voice-Control
SparrowO7
Routes Windows voice commands using speech recognition and an optional LAYA classifier
decision-models-vs-small-llms
pournasserian
Compares Laya and other decision models with small LLMs across six agent decision tasks
aitomaton
sirmmo
Analyzes infrastructure errors and dispatches coding agents based on rules, Laya, Jev, or other LLMs
DecisionOps
ellisgeeker
Certifies and monitors decision-model confidence thresholds to route uncertain predictions for review
laya-vs-local-llm
vyshakhgnair
Compares Laya with a local language model on tool routing and document classification
LAYA-NPU
hellonearthis
Serves Laya-based intent classification and decision routing with Intel NPU acceleration
home-router
cloudeeo
Routes home-assistant requests locally using Laya before running commands, search, or chat
dispatch-system1-router
rafi-seniang
Compares Jev, Laya, and other decision methods in a configurable text-triage router
question-triage-laya
Thaw-Zin-Myo-Aung
Classifies student questions with Laya and routes them to the appropriate learning-assistant workflow
pennywise
1sakshm
Trains a Laya decision model with reinforcement learning to route requests across LLM tiers
laya-gpt
stevenke1981
Uses a local Laya model to select reasoning effort for GPT and resume prematurely ended turns
s1kit
Eliot5566
Tests, calibrates, and routes System One decision models, including Laya, on custom data
laya-toolselect
surajmhaske02
Selects relevant tools for LLM agents with an optional Laya judge
agentic-rag-laya
qaq1902
Tests Laya decision models as optional components in an agentic RAG workflow
hermes-laya-router
axelBrei
Routes home automation requests using Laya decisions and executes supported actions through Home Assistant
Router-LM-Studio
BouchezMarc
Routes prompts to locally hosted LM Studio models based on intent classified by Laya
laya-documentation
P0dp1vass
Documents Laya in Russian with production examples for routing, moderation, prioritization, and HTTP serving
Built-in router preset
The core package ships a question set for small-versus-frontier model routing:
import laya
agent = laya.load("convaiinnovations/laya")
routing = agent.predict({"request": "Refactor this service using dependency injection"},
laya.router_questions())
For LangGraph there is a conditional-edge router with a confidence fallback:
from laya.integrations.langchain import LayaRouter
router = LayaRouter(
criteria={"billing": "invoices, charges", "tech": "bugs, outages"},
confidence_threshold=0.80,
fallback="human_agent",
)
workflow.add_conditional_edges("triage", router)
Note that laya.Router is a different thing: it picks which Laya checkpoint answers, based on script and language.
Measured: Laya vs a reasoning model
laya_router put Laya and GPT-5 nano behind one endpoint and tested 180 labelled requests. Both scored 0.600 route accuracy, at 184 ms p50 for Laya and 6,415 ms for GPT-5 nano. The README also found that rewording the tier descriptions moved accuracy by 21 points.
git clone https://github.com/glukicov/laya_router && cd laya_router
uv sync --all-extras
uv run laya-router serve --backend laya
Projects
- system-one-router: an OpenAI- and Anthropic-compatible Go gateway where Laya or Jev answers topic, complexity, risk and private-data questions, then picks the cheapest adequate model.
- pi-pignon: a pi coding-agent extension that picks a model tier per prompt from a local
laya-serve(about 75 ms on Apple Silicon, per its README) or Jev. - laya-mcp (NVentimiglia): a
route_taskMCP tool for light vs frontier routing.
Caveats
In laya_router, Laya got 97% of small requests right but only 7 of 56 medium ones, so check each tier separately, not just overall accuracy. Write tier descriptions with concrete examples, and send low-confidence answers to a fallback.
Hosted option
layahost runs Laya behind the Jev-compatible /v1/systemone, plus /v1/decide templates for routing, support_intent and priority. Its decision flows chain questions in a visual editor (spam check, then intent, then team) and run the whole flow in one request, asking and billing only the questions on the path taken. Low-confidence answers can take their own branch to a person. layahost is run by the makers of laya.tools.
More ways to use Laya