Reference

This beta feature is currently unavailable. We are moving off a dependency that is no longer open source and onto our own local AI infrastructure, so that every Despia user gets a fully open source, license-compliant experience.

Run language models on-device with one function call. Models load via the device’s native AI acceleration stack. Inference jobs auto-resume across backgrounding. Downloads continue when users close the app.

Installation

Bundle
CDN

npm install despia-intelligence

pnpm add despia-intelligence

yarn add despia-intelligence

import intelligence from 'despia-intelligence';

<script src="https://cdn.jsdelivr.net/npm/despia-intelligence/index.js"></script>

<script type="module">
    import intelligence from 'https://cdn.jsdelivr.net/npm/despia-intelligence/+esm'
</script>

Runtime detection

The SDK resolves runtime state once at import time and exposes it synchronously. Gate every call behind intelligence.runtime.ok so the same code works in a desktop browser preview.

intelligence.runtime.ok       // boolean
intelligence.runtime.status   // 'ready' | 'outdated' | 'unavailable'
intelligence.runtime.message  // string | null

if (!intelligence.runtime.ok) {
  showBanner(intelligence.runtime.message)
  return
}

When ok is false, every API returns a not-ready handle. models.available() resolves to an empty array. The SDK never throws on a missing runtime, so your code can branch cleanly without try/catch.

Run

Fire an inference job and wire callbacks for streaming tokens and the final result.

const call = intelligence.run({
  type:   'text',
  model:  'qwen3-0.6b',
  prompt: 'Summarise this article.',
  system: 'Be concise.',
  stream: true,
}, {
  stream:   (chunk) => output.textContent = chunk,
  complete: (text)  => save(text),
  error:    (err)   => console.error(err.code, err.message),
})

stream(chunk) receives the full accumulated text so far, not a delta. Use el.textContent = chunk (replace), never el.textContent += chunk (append). Appending will produce exponentially duplicated output.

string

required

Routes the call. 'text' is the only enabled value in this release.

string

required

Model id, e.g. 'qwen3-0.6b'. Must be installed first via models.download().

string

required

The user prompt

string

System-level instruction context for the session

boolean

When true, fires stream callbacks as tokens generate

Any extra key on the params object is forwarded to the native layer as-is. Arrays become comma-separated after URL encoding. You do not need to encode values yourself.

intelligence.run({
  type:        'text',
  model:       'qwen3-0.6b',
  prompt:      'Hello.',
  temperature: 0.7,
  top_p:       0.95,
  max_tokens:  256,
}, handler)

Handler callbacks

(chunk: string) => void

Fires for each snapshot. chunk is the full accumulated text so far, not a delta.

(text: string) => void

Fires once when inference finishes. text is the complete response string.

(err: { code, message }) => void

Fires on failure. See error codes.

(intent: object) => void

Optional notification hook. Fires once per active job on focusout. Use for UI affordances or analytics. Resume itself is automatic.

Returns

run() returns a call handle synchronously. The same destructure works whether the runtime is ready or not.

boolean

true when the runtime is ready and the call was queued, false when not

object | null

The original params object, storable and re-firable. null on the not-ready handle.

() => void

Removes this job from the SDK. No further callbacks fire.

const call = intelligence.run(params, handler)

call.intent   // original params object
call.cancel() // drops the job, no further callbacks

Models

Manage the on-device model catalogue. Models are downloaded from Hugging Face into the Despia container and reused across launches.

Available
Installed
Download
Remove

// Full catalogue the runtime can install
const all = await intelligence.models.available()
// [{ id: 'qwen3-0.6b', name: 'Qwen3 0.6B', category: 'text' }, ...]

// Currently downloaded to this device
const installed = await intelligence.models.installed()
const ready = installed.some(m => m.id === 'qwen3-0.6b')

// Fire-and-forget, progress arrives via the callback object
intelligence.models.download('qwen3-0.6b', {
  onStart:    ()    => showDownloadUI(),
  onProgress: (pct) => bar.style.width = pct + '%',
  onEnd:      ()    => hideDownloadUI(),
  onError:    (err) => showError(err),
})

// Remove one model
await intelligence.models.remove('qwen3-0.6b')

// Clear everything (use for a "free up space" button)
await intelligence.models.removeAll()

() => Promise<Model[]>

Full catalogue the runtime can install. Returns [] when runtime.ok is false.

() => Promise<Model[]>

Currently downloaded to this device

(id, { onStart, onProgress, onEnd, onError }) => void

Starts a background download. Fire-and-forget, results arrive via the callback object.

(id: string) => Promise

Remove one model by id

() => Promise

Remove every downloaded model

Download events

Per-call callbacks for the component that started the download. Global events for app-wide state that needs to survive anything, including a force-quit mid-download.

const off = intelligence.on('downloadEnd', (modelId) => markInstalled(modelId))
off() // unsubscribe

intelligence.once('downloadEnd', (modelId) => showFirstDownloadBadge())

(modelId: string) => void

Fires when a download begins

(modelId: string, pct: number) => void

Fires on progress updates. pct is a 0 to 100 integer in both the global event and the per-call onProgress callback.

(modelId: string) => void

Fires when a download completes successfully

(modelId: string, err: object) => void

Fires on download failure

The pattern: session callbacks for in-session UX (a progress bar on the settings page), global events for permanent state (a tab bar badge that needs to survive a force-quit).

Background and return

Inference sessions do not survive backgrounding. The native context is torn down when iOS or Android suspend the WebView. The SDK handles this for you. Every in-flight job is re-fired automatically with the same params and the same handler when the user returns. Just write your code as if backgrounding does not exist. The SDK only re-fires jobs that were genuinely interrupted: jobs that complete normally never re-fire, jobs that error out never re-fire, jobs you explicitly .cancel() never re-fire. Any number of concurrent jobs all resume.

Error codes

Code	Source	Description
`2`	`run`	Missing `id` parameter on the native bridge call
`3`	`run`	Runtime inference error, see `err.message` for detail

intelligence.run(params, {
  error: (err) => {
    if (err.code === 3) {
      console.error('Inference failed:', err.message)
      fallbackToCloud()
    }
  },
})

React hook

import { useState, useEffect, useRef } from 'react'
import intelligence from 'despia-intelligence'

function useInference(model) {
  const [text, setText]       = useState('')
  const [running, setRunning] = useState(false)
  const callRef               = useRef(null)

  const run = (prompt, system) => {
    if (!intelligence.runtime.ok) return
    setText('')
    setRunning(true)
    callRef.current = intelligence.run({
      type: 'text', model, prompt, system, stream: true,
    }, {
      stream:   (chunk) => setText(chunk),
      complete: (full)  => { setText(full); setRunning(false) },
      error:    ()      => setRunning(false),
    })
  }

  useEffect(() => () => callRef.current?.cancel(), [])
  return { text, running, run }
}

Environment check

if (intelligence.runtime.ok) {
  // Use Despia Local AI
} else {
  // Fallback for non-Despia environment
}

The SDK never throws when the runtime is missing. It returns a not-ready handle so your code can branch cleanly. The same code path works in the Despia WebView and in a desktop browser preview.

Resources

NPM Package

despia-intelligence

Introduction

Overview, model selection, and FAQs

GitHub

Source on GitHub

Support

support@despia.com

Deployment

Best Practices

Local Ecosystem

Authentication

Payments

Analytics

Connectivity

Machine Learning

Wallets

Native Features

Editor

Roadblocks

Store Rejections

Platform Support

Installation

Runtime detection

Run

Handler callbacks

Returns

Models

Download events

Background and return

Error codes

React hook

Environment check

Resources

NPM Package

Introduction

GitHub

Support

​Installation

​Runtime detection

​Run

​Handler callbacks

​Returns

​Models

​Download events

​Background and return

​Error codes

​React hook

​Environment check

​Resources

NPM Package

Introduction

GitHub

Support

Installation

Runtime detection

Run

Handler callbacks

Returns

Models

Download events

Background and return

Error codes

React hook

Environment check

Resources