Cursor Core Web Vitals
How real Chrome users experience cursor.com, from Google's Chrome UX Report: the 75th percentile of visits on every device.
Cursor fails Core Web Vitals
Sep 26 – Oct 23, all devices
INP
83 ms
Good
LCP
3.1 s
Needs improvement
CLS
0.00
Good
FCP
2.0 s
Needs improvement
TTFB
628 ms
Good
INP over time
Weekly p75. The dashed line is the 200 ms "good" threshold.
What the numbers say
- Cursor fails Core Web Vitals on Largest Contentful Paint (3.1 s, good is 2.5 s or less).
- Time to first byte is 628 ms, so most of the LCP time is spent after the first byte: loading the main image, fonts or scripts.
- Among 17 developer tools sites we track, Cursor has the 12th fastest LCP.
- INP has held steady around 82 ms over the last 25 weeks.
Keep a site like this from getting slower
Real-user data shows a regression weeks after it ships. Vitalgate runs Lighthouse on every pull request and fails the check when a change breaks budgets like these: LCP ≤ 3.2 s, CLS ≤ 0.10, FCP ≤ 2.0 s.
Other developer tools sites
| Site | Core Web Vitals | LCP | INP | CLS |
|---|---|---|---|---|
| GitHub | Passes | 1.9 s | 87 ms | 0.05 |
| Bitbucket | Fails | 4.0 s | 133 ms | 0.12 |
| Linear | Fails | 3.0 s | 95 ms | 0.10 |
| Jira (Atlassian) | Fails | 4.7 s | 360 ms | 0.11 |
| Sentry | Fails | 2.8 s | 62 ms | 0.00 |
| Postman | Fails | 4.4 s | 116 ms | 0.07 |
| Stack Overflow | Passes | 1.6 s | 61 ms | 0.08 |
| npm | Passes | 1.9 s | 45 ms | 0.01 |
| Docker | Passes | 1.8 s | 110 ms | 0.03 |
| Datadog | Passes | 1.4 s | 78 ms | 0.01 |
| PostHog | Fails | 2.2 s | 384 ms | 0.04 |
| Retool | Passes | 1.5 s | 123 ms | 0.00 |
| Replit | Fails | 6.1 s | 415 ms | 0.10 |
| CodeSandbox | Fails | 4.0 s | 172 ms | 0.04 |
| JetBrains | Fails | 2.2 s | 111 ms | 0.12 |
| Raycast | Passes | 1.8 s | 87 ms | 0.00 |