The all-in-one local AI studio for your desktop: chat, image and video generation and a coding agent in one free, open source app. Windows and Linux. No Docker, no terminal, no cloud required.
TypeScript agentai-chatai-studioall-in-onecoding-agentdesktop-appfluximage-generationlinuxllmlocal-ailocal-llmopen-sourceprivacyself-hostedstable-diffusionuncensoreduncensored-aivideo-generationwindows
import { test, expect, type Page } from '@playwright/test'
import { tauriMockInit, DEFAULT_ASSISTANT_REPLY, DEFAULT_MODEL_NAME } from './support/tauri-mock'
import { openNewChat } from './support/ui'
/**
* P3b acceptance test — fresh onboarding → first chat over the built-in engine.
*
* With no external provider running (the Tauri mock reports Ollama/LM
* Studio/ComfyUI unreachable), a brand-new install must be able to:
* welcome → pick built-in (preselected) → download starter GGUF →
* boot the bundled engine → finish → send a message → get a streamed reply.
*
* The reply text is a fixed marker the mocked engine "generates", so a green
* run proves the whole managed-backend chat path is wired end to end.
*/
async function installMock(page: Page) {
await page.addInitScript(tauriMockInit, {
assistantReply: DEFAULT_ASSISTANT_REPLY,
modelName: DEFAULT_MODEL_NAME,
})
}
test('small GPU keeps the 7B starter visible with a memory advisory', async ({ page }) => {
await installMock(page)
await page.addInitScript(() => {
const bridge = (window as unknown as { __TAURI_INTERNALS__: { invoke: (command: string, args: { url?: string }) => Promise<unknown> } }).__TAURI_INTERNALS__
const invoke = bridge.invoke
bridge.invoke = (command, args) => command === 'proxy_localhost' && args?.url?.includes('/system_stats')
? Promise.resolve(JSON.stringify({ devices: [{ name: 'Test GPU', type: 'cuda', vram_total: 2 * 1024 ** 3 }] }))
: invoke(command, args)
})
await page.goto('/')
await page.getByRole('button', { name: /Get Started/i }).click()
await page.getByRole('button', { name: /Continue/i }).click()
await page.getByRole('button', { name: /Skip for now/i }).click()
const starter = page.getByRole('button', { name: /Qwen 2\.5 7B/i })
await expect(starter).toBeVisible()
await expect(starter).toContainText('Full GPU offload may not fit.')
await expect(page.getByRole('button', { name: /0\.5B/i })).toHaveCount(0)
await expect(page.getByRole('button', { name: /Skip for now/i })).toBeVisible()
})
test('fresh onboarding boots the built-in engine and answers a chat', async ({ page }) => {
await installMock(page)
await page.goto('/')
// welcome
await expect(page.getByRole('button', { name: /Get Started/i })).toBeVisible()
await page.getByRole('button', { name: /Get Started/i }).click()
// backends — built-in engine is preselected, just continue
await expect(page.getByRole('button', { name: /Continue/i })).toBeVisible()
await page.getByRole('button', { name: /Continue/i }).click()
// comfyui — skip (built-in text path doesn't need image gen)
await expect(page.getByRole('button', { name: /Skip for now/i })).toBeVisible()
await page.getByRole('button', { name: /Skip for now/i }).click()
// models — select the starter GGUF, then install. The step auto-advances to
// embeddings once the (mocked) download completes and the engine boots.
await expect(page.getByRole('heading', { name: /Pick a starter model/i })).toBeVisible()
await page.getByRole('button', { name: /Qwen 2\.5 7B/i }).click()
const installBtn = page.getByRole('button', { name: /Install \d+ model/i })
await expect(installBtn).toBeVisible()
await installBtn.click()
// embeddings — skip; onboarding is Ollama-free on the built-in path
await expect(page.getByRole('button', { name: /Skip for now/i })).toBeVisible({ timeout: 30_000 })
const downloads = await page.evaluate(() => (window as unknown as { __E2E_DL_CALLS__: unknown[] }).__E2E_DL_CALLS__)
expect(downloads).toContainEqual(expect.objectContaining({
filename: 'Qwen2.5-7B-Instruct-abliterated-v2.Q4_K_M.gguf',
expectedBytes: 4683073920,
expectedSha256: 'da9272f09cd51d27aaadf37c1a2a3e079763682688aa1601d7406062b0caa5aa',
}))
await page.getByRole('button', { name: /Skip for now/i }).click()
// done — finish sets onboardingDone and reveals the chat UI
await expect(page.getByRole('button', { name: /Get Started/i })).toBeVisible()
await page.getByRole('button', { name: /Get Started/i }).click()
// chat — open a fresh conversation, which reveals the composer
await openNewChat(page)
const composer = page.locator('textarea').first()
await composer.fill('ping the built-in engine')
await page.getByRole('button', { name: /Send message/i }).click()
// streamed assistant reply from the mocked built-in engine
await expect(page.getByText(/PONG_BUILTIN_OK/)).toBeVisible({ timeout: 20_000 })
})