mlcommons.org

Ranks Overall #1,569 Built Awesome #1,039

Data from Oct 4, 2026, 10:55 AM UTC 4d old Measured once a week · 8 runs since Aug 21, 2026

Elsewhere: PageSpeed Insights · Calibre · Treo · CrUX Vis · Tame the Bots · Scoring Calculator · Wayback Machine

The measuring machine was faster than most in this project.

Lighthouse benchmark index was 3,837, 54% faster than this project's median of 2,486. CPU-bound metrics — TBT, TTI, main thread work — are affected by this and may have moved without the site changing at all.

Lab data is measured using Lighthouse, Mobile mode unless otherwise specified. Field data is reported from Chrome UX Report (using p75 measurements).

Stack

Hosted by Cloudflare

Lighthouse Scores LAB Simulated

Performance

↑ +10 (+17.2%)

Accessibility

no change

Best Practices

no change

SEO

no change

Metrics LAB FIELD

Core Web Vitals: passing 3 of 3 assessed from real Chrome users.

Field figures are the p75 across real Chrome users over a 28-day trailing window (2026-09-05 → 2026-10-02), url-level.

Metric	Lab Simulated	Field
LCP CWV Largest Contentful Paint	13.76 s poor	↑ +1166 ms (+9.3%) 12.60 s – 16.89 s	2.39 s good	79.8% 13.3% 6.9%
INP CWV Interaction to Next Paint — field only, no lab run can measure it	n/a	—	118 ms good	89.1% 6.3% 4.7%
CLS CWV Cumulative Layout Shift	0 good	→ no change 0 – 0	0 good	100% 0% 0%
TBT Total Blocking Time	95 ms good	↓ -275 ms (-74.3%) significant 95 ms – 436 ms	n/a	—
FCP First Contentful Paint	2.55 s needs work	↓ -818 ms (-24.3%) 2.55 s – 3.61 s	1.15 s good	87.5% 8.1% 4.5%
Speed Index	4.59 s needs work	↑ +641 ms (+16.2%) 3.51 s – 5.95 s	n/a	—
TTFB Time to First Byte	657 ms good	↑ +25 ms (+4%) 500 ms – 683 ms	574 ms good	83.2% 11.1% 5.6%

Metric

Lab Simulated

Field

Value

Change / range

Trend

p75

Distribution

LCP CWV Largest Contentful Paint

13.76 s poor

↑ +1166 ms (+9.3%) 12.60 s – 16.89 s

2.39 s good

79.8% 13.3% 6.9%

INP CWV Interaction to Next Paint — field only, no lab run can measure it

n/a

—

118 ms good

89.1% 6.3% 4.7%

CLS CWV Cumulative Layout Shift

0 good

→ no change 0 – 0

0 good

100% 0% 0%

TBT Total Blocking Time

95 ms good

↓ -275 ms (-74.3%) significant 95 ms – 436 ms

n/a

—

FCP First Contentful Paint

2.55 s needs work

↓ -818 ms (-24.3%) 2.55 s – 3.61 s

1.15 s good

87.5% 8.1% 4.5%

Speed Index

4.59 s needs work

↑ +641 ms (+16.2%) 3.51 s – 5.95 s

n/a

—

TTFB Time to First Byte

657 ms good

↑ +25 ms (+4%) 500 ms – 683 ms

574 ms good

83.2% 11.1% 5.6%

LCP Subparts LAB Unthrottled

The subparts attribute LCP to your server, resource discovery, download, or rendering. Note that the Lab LCP shown above is a simulated throttled estimate.

element render delay 931 ms

resource load duration 201 ms

time to first byte 28 ms

resource load delay 10 ms

LCP LAB Simulated vs LCP FIELD

Lab 13.76 s

Field p75 2.39 s

The lab LCP is 5.76× the field p75 — real users likely have above-average hardware or network connectivity.

Page weight

Metric	Value	Change / range
Page Weight	3.03 MB	↑ +467.3 kB (+18.8%) 2.55 MB – 4.34 MB
Requests	93	→ no change 93 – 95
Third-party Weight	249 kB	↑ +1.7 kB (+0.7%) 243 kB – 249 kB
Third-party CPU	55 ms	↓ -55 ms (-50.1%) significant 55 ms – 130 ms
Unused JS	96.9 kB	↓ -1.8 kB (-1.8%) 95.8 kB – 98.8 kB
Unused CSS	56.7 kB	↑ +3 B 56.7 kB – 56.7 kB

Metric

Value

Change / range

Trend

Page Weight

3.03 MB

↑ +467.3 kB (+18.8%) 2.55 MB – 4.34 MB

Requests

→ no change 93 – 95

Third-party Weight

249 kB

↑ +1.7 kB (+0.7%) 243 kB – 249 kB

Third-party CPU

55 ms

↓ -55 ms (-50.1%) significant 55 ms – 130 ms

Unused JS

96.9 kB

↓ -1.8 kB (-1.8%) 95.8 kB – 98.8 kB

Unused CSS

56.7 kB

↑ +3 B 56.7 kB – 56.7 kB

By resource type

Type	Share	Size	Requests
image	58%	1.76 MB	46
other	19.7%	598 kB	8
script	13.6%	413 kB	22
document	4.3%	131 kB	2
stylesheet	2.4%	73.1 kB	6
font	1.9%	57.7 kB	3
Total		3.03 MB	93

Third parties 3 entities

Entity	Transfer	Size	Main thread Simulated
Google Tag Manager	script	161 kB	46 ms
Google Fonts	font	59.6 kB	0 ms
cloudflare.com	script	28.3 kB	9 ms
Total		249 kB	55 ms

Entity

Transfer

Size

Main thread Simulated

Google Tag Manager

script

161 kB

46 ms

Google Fonts

font

59.6 kB

0 ms

cloudflare.com

script

28.3 kB

9 ms

Total

249 kB

55 ms

Page health

Metric	Value	Change / range
Main Thread Work	1.59 s	↓ -773 ms (-32.7%) significant 1.59 s – 2.47 s
Long Tasks	4	↓ -2 (-33.3%) 4 – 10
DOM Elements	780	↑ +4 (+0.5%) 772 – 780
Axe Violations Violating nodes from a full Axe CLI run (more than Lighthouse reports)	12	→ no change 12 – 12
A11y Failures	4	→ no change 4 – 4
A11y Failing Nodes	10	→ no change 10 – 10

Metric

Value

Change / range

Trend

Main Thread Work

1.59 s

↓ -773 ms (-32.7%) significant 1.59 s – 2.47 s

Long Tasks

↓ -2 (-33.3%) 4 – 10

DOM Elements

780

↑ +4 (+0.5%) 772 – 780

Axe Violations Violating nodes from a full Axe CLI run (more than Lighthouse reports)

→ no change 12 – 12

A11y Failures

→ no change 4 – 4

A11y Failing Nodes

→ no change 10 – 10

Main thread work

Task	Share	Duration Simulated
other	41%	649 ms
script evaluation	30%	477 ms
style layout	12%	190 ms
script parse compile	7%	109 ms
parse HTML	5%	82 ms
paint composite render	5%	75 ms
garbage collection	1%	13 ms
Total		1.59 s

Task

Time

Duration Simulated

other

41%

649 ms

script evaluation

30%

477 ms

style layout

12%

190 ms

script parse compile

109 ms

parse HTML

82 ms

paint composite render

75 ms

garbage collection

13 ms

Total

1.59 s

Axe violations axe-core 4.13.0, full rule set

12 violating nodes across 6 rules, 961 passing. This is a standalone Axe CLI run and is stricter than the subset Lighthouse already includes.

serious link-name — Links must have discernible text 5 nodes

Offending elements

article:nth-child(1) > figure > a

<a href="https://mlcommons.org/2026/10/agent-privacy-risk-taxonomy-v0-1/">

article:nth-child(2) > figure > a

<a href="https://mlcommons.org/2026/10/dlrmv4-training/">

article:nth-child(3) > figure > a

<a href="https://mlcommons.org/2026/09/mlperf-training-llm-post-training/">

article:nth-child(4) > figure > a

<a href="https://mlcommons.org/2026/09/join-airis-biomedical-ai/">

article:nth-child(5) > figure > a

<a href="https://mlcommons.org/2026/09/chairs-mlperf-inference-v6-1/">

critical button-name — Buttons must have discernible text 3 nodes
Offending elements
1. .burger
```
<button class="burger">
```
2. .nav-buckets__primary > .site-search[role="search"][method="get"] > .search-field > button[type="submit"]
```
<button type="submit">
```
3. .wrapper > .site-search[role="search"][method="get"] > .search-field > button[type="submit"]
```
<button type="submit">
```
moderate heading-order — Heading levels should only increase by one 1 node
Offending element
1. .wp-block-column.is-layout-flow.wp-block-column-is-layout-flow:nth-child(1) > h4
```
<h4 class="wp-block-heading">Performance Benchmarks</h4>
```
moderate landmark-unique — Landmarks should have a unique role or role/label/title (i.e. accessible name) combination 1 node
Offending element
1. .nav-buckets__primary > .site-search[role="search"][method="get"]
```
<form role="search" method="get" class="site-search" action="https://mlcommons.org/">
```
moderate region — All page content should be contained by landmarks 1 node
Offending element
1. .skip-link
```
<a class="skip-link screen-reader-text" href="#primary">Skip to content</a>
```
moderate skip-link — The skip-link target should exist and be focusable 1 node
Offending element
1. .skip-link
```
<a class="skip-link screen-reader-text" href="#primary">Skip to content</a>
```

Accessibility failures 4 of 30 audits

link-name — Links do not have a discernible name 5 nodes
button-name — Buttons do not have an accessible name 3 nodes
heading-order — Heading elements are not in a sequentially-descending order 1 node
skip-link — Skip links are not focusable. 1 node

Best Practices failures 1 of 19 audits

errors-in-console — Browser errors were logged to the console 1 item

SEO failures 1 of 10 audits

link-text — Links do not have descriptive text 3 items

Hygiene

HTTPS HSTS CSP Clickjacking protection Origin isolation Trusted Types bfcache h2 1 console error

Measurement log

Every stored run for this URL, newest first.

When	Perf	LCP	TBT	Weight	Runs	Spread	Bench	Took	Archive
Oct 4, 2026, 10:55 AM UTC	68	13.76 s	95 ms	3.03 MB	3/3	±4	3,837 +54%	37.6s	Wayback
Sep 27, 2026, 6:16 AM UTC	58	12.60 s	370 ms	2.55 MB	3/3	±27	2,489 0%	45.2s	Wayback
Sep 20, 2026, 12:40 AM UTC	59	12.98 s	436 ms	2.83 MB	3/3	±20	2,453 -1%	46.2s	Wayback
Sep 12, 2026, 11:45 PM UTC	62	15.03 s	306 ms	3.63 MB	3/3	±23	2,462 -1%	57.4s	Wayback
Sep 5, 2026, 10:38 PM UTC	66	14.71 s	163 ms	3.60 MB	3/3	±7	2,910 +17%	49.4s	Wayback
Sep 5, 2026, 8:05 PM UTC	65	15.68 s	100 ms	3.64 MB	3/3	±1	3,408 +37%	42.6s	Wayback
Aug 29, 2026, 7:43 PM UTC	60	16.01 s	354 ms	4.01 MB	3/3	±0	2,455 -1%	36.5s	Wayback
Aug 21, 2026, 4:44 AM UTC	59	16.89 s	260 ms	4.34 MB	3/3	±6	2,469 -1%	36.3s	Wayback

When

Scores

Perf

LCP

CLS

TBT

Weight

Runs

Spread

Bench

Took

Embed this score

Use this web component to show mlcommons.org's scores on any website, updated whenever this instance re-measures. Don’t forget to bundle the speedlify2-score.js component code into your site!

<speedlify2-score speedlify-url="https://www.speedlify.dev/" url="https://mlcommons.org/"></speedlify2-score>

Stack

Lighthouse Scores LAB 🤖 Simulated

Metrics LAB FIELD

LCP Subparts LAB 🐇 Unthrottled

LCP LAB 🤖 Simulated vs LCP FIELD

Page weight

By resource type

Third parties 3 entities

Page health

Main thread work

Axe violations axe-core 4.13.0, full rule set

Accessibility failures 4 of 30 audits

Best Practices failures 1 of 19 audits

SEO failures 1 of 10 audits

Hygiene

Measurement log

Embed this score

Lighthouse Scores LAB Simulated

LCP Subparts LAB Unthrottled

LCP LAB Simulated vs LCP FIELD