JS libraries impact

Performance impact by framework provider.

Field data PhoneDesktopAll Scope All sites Q1 2026 edition · Desktop field outcomes
Metric LCP INP CLS
1

At a glance the headline numbers for JS libraries impact

Performance impact by framework provider.

20
Providers ranked
After min-sites filter
142,710
Sites in sample
Combined across all items
1.4s
Best LCP (p75)
Turbo/Hotwire
The State of Web Vitals · Q1 2026 · 94,910 sites · desktop field datacorewebvitals.io/state-of-cwv
2

The ranking sorted by LCP at p75, fastest first

# Provider LCP p75 Passing Sites
1 Turbo/Hotwire 1.4s 95.9%
258
2 Stimulus/Hotwire 1.8s 93.5%
1,536
3 Flickity 2.0s 87.6%
2,891
4 Next.js 2.0s 86.4%
2,085
5 jQuery 2.0s 85.4%
67,906
6 Slick Carousel 2.0s 85.9%
7,462
7 Lazysizes 2.1s 85.9%
6,844
8 Owl Carousel 2.1s 84.3%
4,359
9 Vue.js 2.1s 83.0%
5,677
10 React 2.1s 81.4%
235
11 Lit 2.1s 83.7%
5,243
12 Swiper 2.2s 81.8%
15,692
13 Moment.js 2.2s 80.1%
4,065
14 GSAP 2.3s 79.3%
2,573
15 Ember.js 2.3s 79.5%
67
16 Lodash 2.4s 77.6%
13,200
17 Nuxt 2.5s 76.3%
647
18 Backbone.js 2.5s 74.1%
818
19 Angular 2.7s 69.9%
1,036
20 Ionic 3.1s 60.5%
116
The State of Web Vitals · Q1 2026 · 94,910 sites · desktop field datacorewebvitals.io/state-of-cwv
3

Passing LCP per provider which group passes the LCP most often

ProviderSitesPassing LCPp75
Turbo/Hotwire 0.3% 258 96% 1.4s
Stimulus/Hotwire 1.6% 1,536 94% 1.8s
Marko 0% 12 91% 2.2s
Flickity 3% 2,891 88% 2.0s
Next.js 2.2% 2,085 86% 2.0s
Slick Carousel 7.9% 7,462 86% 2.0s
lazysizes 7.2% 6,844 86% 2.1s
jQuery 71.5% 67,906 85% 2.0s
Owl Carousel 4.6% 4,359 84% 2.1s
Lit 5.5% 5,243 84% 2.1s
Vue.js 6% 5,677 83% 2.1s
Swiper 16.5% 15,692 82% 2.2s
React 0.2% 235 81% 2.1s
Moment.js 4.3% 4,065 80% 2.2s
Ember.js 0.1% 67 80% 2.3s
GSAP 2.7% 2,573 79% 2.3s
Polyfill.io 0% 46 78% 2.1s
Lodash 13.9% 13,200 78% 2.4s
Nuxt.js 0.7% 647 76% 2.5s
Backbone.js 0.9% 818 74% 2.5s
Angular 1.1% 1,036 70% 2.7s
Ionic 0.1% 116 61% 3.1s
Good Needs Improvement Poor Sorted best-passing first · median colored by its own rating · pass = good LCP (2.5s at p75) · one value per site

96% of Turbo/Hotwire sites pass LCP. Ionic trails 35 points behind, leaving 39% of its sites failing. computed

The State of Web Vitals · Q1 2026 · 94,910 sites · desktop field datacorewebvitals.io/state-of-cwv
4

All five vitals at once the whole category without toggling - cell is the pass rate, small number the p75

LCP
INP
CLS
FCP
TTFB
jQuery 71.5%
852.0s
9962ms
780.09
801.6s
571.2s
Swiper 16.5%
822.2s
9967ms
790.09
771.7s
521.4s
Lodash 13.9%
782.4s
9869ms
720.12
731.9s
511.4s
Slick Carousel 7.9%
862.0s
9964ms
720.12
811.6s
551.2s
lazysizes 7.2%
862.1s
9967ms
760.10
811.6s
571.2s
Vue.js 6%
832.1s
9976ms
720.12
811.6s
611.1s
Lit 5.5%
842.1s
9975ms
740.11
811.6s
641.0s
Owl Carousel 4.6%
842.1s
9962ms
690.13
801.6s
521.3s
Moment.js 4.3%
802.2s
9966ms
720.12
761.7s
571.3s
Flickity 3%
882.0s
10067ms
820.07
841.5s
651.1s
GSAP 2.7%
792.3s
10065ms
820.08
761.7s
521.4s
Next.js 2.2%
862.0s
95106ms
750.10
881.4s
76773ms
Stimulus/Hotwire 1.6%
941.8s
9969ms
840.08
921.3s
81706ms
Angular 1.1%
702.7s
97111ms
440.33
791.7s
85539ms
Backbone.js 0.9%
742.5s
9967ms
720.13
682.0s
461.5s
Nuxt.js 0.7%
762.5s
98100ms
650.16
781.7s
641.0s
Turbo/Hotwire 0.3%
961.4s
9980ms
930.03
931.1s
83692ms
React 0.2%
812.1s
91111ms
590.19
841.6s
651.1s
Ionic 0.1%
613.1s
94106ms
680.12
582.2s
671.3s
Ember.js 0.1%
802.3s
9782ms
720.15
731.9s
68945ms
Polyfill.io 0%
782.1s
10074ms
790.09
841.4s
69889ms
Marko 0%
912.2s
10071ms
1000.01
551.9s
01.5s
60%95%+ passing Cell: pass rate, small number = p75 · faded rows: under 100 sites

One row per provider, one column per vital - the cell is the share of sites passing, the small number the p75. No toggling needed to see where the category actually differs.

Turbo/Hotwire leads on LCP: 96% of its sites pass. Ionic trails at 61%. computed

The State of Web Vitals · Q1 2026 · 94,910 sites · desktop field datacorewebvitals.io/state-of-cwv
Related signals A/B testing (10) → Personalization (5) → E-commerce (14) → CDN (15) → Chrome field data from 94,910 sites, representing millions of real page loads · How we measured