> For the complete documentation index, see [llms.txt](https://unsloth.ai/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://unsloth.ai/docs/models/decision-laya.md).
# Run and Serve Decision Models Locally with Unsloth: Laya + Jev API
Serve and run decision models locally with Unsloth.
Run and serve decision models like **Laya**, an open Jev alternative, on your local **Mac**, **Windows** and **Linux** device with [Unsloth](https://github.com/unslothai/unsloth). Decision models turns text inputs into structured decisions your apps can act on - from routing support tickets, filtering emails to automated tags and classifications.
You can use the Unsloth GUI interface to expose Laya through a TypeSafe-compatible Jev API, so existing Jev integrations can work by pointing them at your local server.
All data stays on your device and **CPU**, **unified memory**, and **GPU** setups are all supported.
Send the decision model pieces of text, define the decisions you need (yes/no, choices, or scores), and get back calibrated probabilities.
### 🖥️ Recommended Settings
| Model | Minimum RAM | Size | Best for |
| -------------------------- | ----------- | ------ | --------------------------------------------------------------------------- |
| **Multilingual** (default) | 4GB RAM | 678 MB | Most use-cases. 100+ languages, 1024 token context. |
| **English** | 5GB RAM | 846 MB | English only. Larger encoder, 512 token context. |
| **Typed decisions** | 5GB RAM | 846 MB | English only. Invoices, security incidents, customer service, agent traces. |
CPU is the default. Switch **Run on** GPU for faster answers. The GPU keeps the model in memory until you press **Unload** or restart Unsloth. If the GPU runs out of memory, Laya falls back to CPU and keeps answering.
### ⚡ Quickstart
{% stepper %}
{% step %}
#### **Setup Unsloth**
The easiest way to get started is by downloading the [Unsloth Desktop app](/docs/desktop.md). Works on [macOS](/docs/get-started/install/mac.md), [Windows](/docs/get-started/install/windows-installation.md), and [Linux](/docs/get-started/install/linux.md).
Download Unsloth
* :apple: [Download for macOS](https://unsloth.ai/download/mac)
* :windows: [Download for Windows](https://unsloth.ai/download/windows)
* :linux: [Download for Linux](https://unsloth.ai/download/linux)
Or, if you prefer to install manually:
MacOS, Linux, WSL:
```bash
curl -fsSL https://unsloth.ai/install.sh | sh
```
Windows PowerShell:
```bash
irm https://unsloth.ai/install.ps1 | iex
```
{% endstep %}
{% step %}
#### Setup Laya
Go to Settings → API → **Decision API** and turn on **Serve requests**. The default Laya model (678 MB) will then start to download after confirming if you want to start the download.
Your API key is on the same page under Access tokens. Or turn on **Keyless API access** → **Chat and inference** and skip the key on localhost.
{% endstep %}
{% step %}
#### Send a decision request
It's time to send a decision request. In this example, we classify a support ticket, detect refund requests, and score urgency.
```bash
curl http://localhost:8888/v1/systemone \
-H "Authorization: Bearer sk-unsloth-YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "laya",
"state": "Hi, I was charged twice for my March invoice (#4411). Please refund the duplicate today.",
"questions": {
"team": {
"type": "choice",
"instructions": "Which team should handle this?",
"criteria": {
"billing": "invoices, payments, refunds",
"technical": "bugs, outages, errors",
"sales": "pricing, new plans",
"other": "everything else"
}
},
"refund": {"type": "noul", "instructions": "Does the customer ask for a refund?"},
"urgency": {"type": "score", "instructions": "How urgent is this?", "criteria": ["not urgent", "soon", "today"]}
}
}'
```
You get back:
```json
{
"model": "laya-multilingual",
"answers": {
"team": {"type": "choice", "choice": "billing", "confidence": 1.0, "probabilities": {"billing": 1.0, "technical": 0.0, "sales": 0.0, "other": 0.0}},
"refund": {"type": "noul", "noul": 0.9794},
"urgency": {"type": "score", "score": 1.9005, "confidence": 0.698, "legend": {"0": "not urgent", "1": "soon", "2": "today"}, "probabilities": {"0": 0.004, "1": 0.0915, "2": 0.9045}}
},
"usage": {"input_tokens": 172, "output_tokens": 0}
}
```
The first request loads the model, which takes 10 to 20 seconds. After that, requests take well under a second on most CPUs.
`state` can be a string or any JSON. `model` accepts `laya`, `default`, or `jev-latest` for the model you picked in settings, or `laya-multilingual`, `laya-english`, `laya-typed-decisions` to name one.
{% endstep %}
{% endstepper %}
### ❓ Question types
| Type | Use it for | You get back |
| -------- | ------------------ | --------------------------------------------------- |
| `noul` | Yes or no | The probability of yes, from 0 to 1 |
| `choice` | Picking one option | The most likely option, and a probability for each |
| `score` | Rating on a scale | Where the text falls on your scale, counting from 0 |
`noul` means yes/no. It returns the probability of “yes”: `0.98` means very likely, `0.5` means equally likely either way.
`choice` and `score` also return `confidence`. Close to `1` means one answer stands out. Close to `0` means the probabilities are spread evenly.
{% columns %}
{% column %}
To act on a choice, use its value in `probabilities`, not `confidence`. For example, only file a bug report when `technical` is above `0.8`. Laya and Jev calculate confidence differently, so don’t reuse your Jev confidence thresholds.
Each request supports up to **64 questions**, with up to **255 options per choice** and **10 levels per score**.
{% endcolumn %}
{% column %}
Real time packing demo powered by local Laya through Unsloth’s Decision API: suitcase items update as you type and change your travel plans
{% endcolumn %}
{% endcolumns %}
### 🔁 Migrating from Jev
Install the SDK with `pip install typesafe-sdk`, set two environment variables, and your existing code talks to Unsloth:
```bash
export TYPESAFE_BASE_URL=http://localhost:8888
export TYPESAFE_API_KEY=sk-unsloth-YOUR_KEY
```
The SDK's default model, `jev-latest`, maps to whichever model you picked in Settings.
Two things differ from Jev. `confidence` uses a different formula, so a threshold tuned on Jev won't transfer; threshold on `probabilities` instead. And options in a choice share a fixed token budget, so past about 20 described options the labels get trimmed.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://unsloth.ai/docs/models/decision-laya.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
