mirror of
https://github.com/mudler/LocalAI.git
synced 2026-11-02 10:34:46 -05:00
localai-org-maint-botandlocalai-org-maint-bot
1ac2ecb2df
Streaming chunks can contain multiple tokens. Use reported completion totals for the final end-to-end average and label live rates as estimates. Count each client tool turn once and include reasoning in the total. Assisted-by: Codex:gpt-6 Co-authored-by: localai-org-maint-bot <306269227+localai-org-maint-bot@users.noreply.github.com>
49 lines
2.3 KiB
JavaScript
49 lines
2.3 KiB
JavaScript
import { test, expect } from './coverage-fixtures.js'
|
|
import { chatOf, mockChat, openChat, controlStream } from './chat-fixtures.js'
|
|
|
|
async function start(page) {
|
|
await mockChat(page)
|
|
await controlStream(page)
|
|
await openChat(page, [chatOf('c1', 'Throughput', 'qwen3-8b', [])])
|
|
// Only Date is fixed: browser timers and React continue running normally.
|
|
await page.clock.setFixedTime(new Date('2026-10-26T12:00:00Z'))
|
|
await page.getByTestId('chat-input').fill('hello')
|
|
await page.getByTestId('chat-send').click()
|
|
await expect.poll(() => page.evaluate(() => window.__sse.calls)).toBe(1)
|
|
}
|
|
|
|
async function atTwoSeconds(page, usage) {
|
|
await page.clock.setFixedTime(new Date('2026-10-26T12:00:02Z'))
|
|
await page.evaluate((usage) => {
|
|
window.__sse.push({ choices: [{ delta: { reasoning: 'thinking about the answer' } }] })
|
|
window.__sse.push({ choices: [{ delta: { content: 'The answer has several tokens.' } }] })
|
|
if (usage) window.__sse.push({ choices: [], usage })
|
|
window.__sse.end()
|
|
}, usage)
|
|
}
|
|
|
|
test('chat throughput uses completion usage, including reasoning, over request duration', async ({ page }) => {
|
|
await start(page)
|
|
await atTwoSeconds(page, { prompt_tokens: 10, completion_tokens: 100, total_tokens: 110,
|
|
completion_tokens_details: { reasoning_tokens: 80 } })
|
|
await expect(page.getByTestId('chat-foot')).toContainText('50 tok/s · end-to-end')
|
|
await expect(page.getByTestId('chat-foot')).not.toContainText('estimated')
|
|
await expect(page.getByTestId('chat-foot')).not.toContainText('Peak')
|
|
})
|
|
|
|
test('chat throughput labels live and missing-usage rates as estimates', async ({ page }) => {
|
|
await start(page)
|
|
await page.clock.setFixedTime(new Date('2026-10-26T12:00:01Z'))
|
|
await page.evaluate(() => window.__sse.push({ choices: [{ delta: { content: '12345678' } }] }))
|
|
await expect(page.getByTestId('chat-foot')).toContainText('2 tok/s · estimated')
|
|
await page.clock.setFixedTime(new Date('2026-10-26T12:00:02Z'))
|
|
await page.evaluate(() => window.__sse.end())
|
|
await expect(page.getByTestId('chat-foot')).toContainText('1 tok/s · estimated')
|
|
})
|
|
|
|
test('chat throughput accepts a reported zero completion count', async ({ page }) => {
|
|
await start(page)
|
|
await atTwoSeconds(page, { prompt_tokens: 10, completion_tokens: 0, total_tokens: 10 })
|
|
await expect(page.getByTestId('chat-foot')).toContainText('0 tok/s · end-to-end')
|
|
})
|