---
title: 'Is PHP 8.5 faster than previous versions?'
source: 'https://youtube.com/watch?v=kb_8u0jB6OA'
video_id: 'kb_8u0jB6OA'
date: 2026-07-28
duration_sec: 597
---

# Is PHP 8.5 faster than previous versions?

> Source: [Is PHP 8.5 faster than previous versions?](https://youtube.com/watch?v=kb_8u0jB6OA)

## Summary

PHP 8.5 was recently released, and benchmarks comparing it to PHP 8.2–8.4 on Symfony, Laravel, and WordPress show no significant performance improvement. The video explains the benchmark setup, results, and why general speed gains are hard at this point.

### Key Points

- **PHP 8.5 release and benchmark motivation** [0:00] — PHP 8.5 was released last week. The team benchmarked it against previous versions using Symfony, Laravel, and WordPress demos.
- **Benchmark setup** [0:52] — Used a Hetzner machine with 8 dedicated virtual cores, Debian 13 Trixie, MySQL/SQLite, and PHP 8.5 RC3 from sury packages. JIT disabled, PHP-FPM with 17 static workers.
- **Project versions and methodology** [1:42] — Laravel 12, Symfony 7.3, WordPress 6.8. Two tests: fixed requests per second (response time) and fixed concurrency (throughput), each run for 60 seconds.
- **Simple test setup rationale** [2:33] — Chose default PHP configurations to reflect typical usage, as 95% of users run vanilla setups.
- **Symfony results** [3:12] — Response times and throughput are nearly identical across PHP 8.2–8.5, within margin of error.
- **Laravel results** [4:44] — Similar performance across versions; no significant difference in response time or throughput.
- **WordPress results** [5:46] — PHP 8.3–8.5 show slightly better throughput than PHP 7.4, but no difference among PHP 8 versions.
- **Conclusion and advice** [7:14] — PHP 8.5 is not faster than 8.4 or earlier. Performance improvements now require application-level optimizations or using new APIs like persistent curl state.

### Conclusion

PHP 8.5 offers no general speed increase over PHP 8.4 or earlier versions. To improve performance, developers should focus on application-level changes, such as optimizing SQL queries, using modern APIs, or adopting application servers like FrankenPHP.

## Transcript

PHP 8.5 was just released last week and
naturally you are as curious as I am if
it's faster than previous versions of
PHP. My colleague Tim set out to test
this with a benchmark for PHP 8.4 and
comparing a Symfony Laravel and a
WordPress demo applications against
previous versions of PHP.
Can you expect a big performance
increase like from PHP 5 to 7 or is
there no noticeable increase? In this
video, I'm going to discuss our
findings. Mo, I am Benjamin and my team
and I work on PHP performance 24/7 for
the last 10 years, helping thousands of
developers along the way. Let's start
discussing the benchmark results that
Tim collected. You find all of them in
our blog post. The link is in the
description. First, the setup is always
important. Uh the blog post lists
everything that went into this
benchmark. So, first you will find all
the code that we use for infrastructure
provisioning and the demo applications.
We used a Hetsa machine with eight
dedicated virtual cores. We are using uh
the Debian 13 Trixie release uh for a
database MySQL or SQLite depending on
the demo. And the uh one important
factor is for PHP we used um the suri um
packages built for Debian um including
PHP 8.5 release candidate number three
for the PHP setup we didn't change a lot
uh the JIT is not enabled and uh PHP FPM
is configured with a static pool of 17
workers.
The for the project versions we have
Laravel version 12, Symphfony 7.3,
WordPress 6.8 and we used Va for um uh
doing performing the benchmark testing.
As for our methodology, what we ran here
is two different kinds of benchmarks. So
we tested with a fixed request per
second number so that we can focus on
the response time of PHP um with this
fixed number of requests. And the second
test we did is with a fixed level of
concurrency to focus on the sort of
throughput. How many requests per second
did we achieve by running this fixed
level of concurrency. Uh both tests run
for about 60 seconds or for 60 seconds
and then we compile these numbers and uh
generate charts from them.
We try to make the test setup as simple
as possible because we usually expect
that you don't reconfigure PHP too um
broadly. So what we didn't want to test
is um PHP versions that are
self-compiled or um use very complex
setup of the demo applications or PHP
itself because from experience
um 95% of you will be running PHP on um
more vanilla uh configurations the
default configurations.
So um we wanted to replicate this kind
of running PHP. Let's go through the
results one by one. We are starting with
Symfony and we are looking at the test
scenario where we run 100 requests per
second a fixed number of requests every
second and then we are looking at uh the
response times that we see from the uh
demo application. So for a symfony
application we can see across versions
PHP 8.2 2 to 8.5 that the response times
are slightly slower for PHP 8.5 compared
to 8.2 or 8.3 for example in the mean.
Uh also like super slightly slower in
the median 95 percentile and 99
percentile.
This is not statistically
uh significant. So um it is within the
margin of error. um the numbers could
statistically be the same. So there's
really no difference in the performance
between these versions if we um perform
this kind of test where we run a fixed
number of requests um against the
Symfony application every second. For
the second test for Symfony uh running
15 concurrent users to find out how many
requests is the application being able
to process in a fixed time frame. we see
there um across all PHP versions it's
around 500 requests and uh again it's
within the margin of error so um it's
not clear which version here is faster
or not they all seem to be equally fast
same for the Laravel case uh first for
the test with 100 requests per second uh
testing the response time we can see uh
in the mean median 95 and 99 percentile
are roughly the same there seems to be
one outlier for PHP 8.3 in the 99th
percentile but this is also not very big
and um selfcorrected itself in the next
so in PHP 8.3 so in general again here I
would say the performance for PHP 8.5 is
the same than the previous versions 82
to 8.4 for for Laravel case with the
scenario of 15 concurrent users making
as many requests as possible in 60
seconds. We can also see here around 850
requests for this Laravel demo
application and it's the same uh
regardless of the PHP version. So again
it doesn't seem to be that PHP 8.5 is
generally faster than previous versions.
As a third case, we've took WordPress um
specifically also because it's possible
to run WordPress still with PHP 7.4. So
here we can test is there a performance
difference between PHP 7.4 and all the
PHP 8 versions including 8.5. And for
the test with 100 requests per second,
we can see the performance is not really
um very different between these
versions. um there is within the margin
of error no real um improvement between
all these versions.
So we can conclude that PHP 8.5 is also
not faster for WordPress applications
compared to previous versions. For a
test with 15 concurrent users with
WordPress, we this is the only case
where we can see a small difference that
there um is outside of the margin of
error. So we can see for PHP 8.3 4 and 5
the number of requests that we can uh
achieve in a fixed time frame is a
little bit larger than for PHP 7.4. So
um there is this difference between PHP
8 and PHP 7. But again PHP 8.5 is not um
be able to handle more requests per uh
time frame than PHP 8.4 or 8.3.
What do these results mean in general?
Uh so we found out PHP 8.5 is not faster
than PHP 8.4 or previous versions than
PHP. So uh the result that we already
saw last year when we benchmark PHP 8.4
is still true. PHP is at a very high
level of performance for a dynamics
language and achieving general
improvements is really hard at that
point. Uh to achieve changes you
actually need to look at the application
and runtime level of PHP. What you can
obviously do is check the performance of
your PHP application in uh specifically
are there SQL queries that are slow? Are
you using APIs the wrong way? Maybe
there are more modern APIs in PHP now
that you that weren't available
previously.
Uh we published a blog post on
improvements in PHP 8.5 that will
significantly improve the performance of
specific use cases. But uh you need to
find if you are actually using PHP and
using these APIs and are benefiting uh
from them. One thing I want to note here
is for example the persistent storage of
curl state across um requests. Uh this
uh will allow applications that talk to
third party APIs to um improve the
performance quite significantly with
8.5.
You can find a blog post in the
description uh on this um on on the
specific PHP 8.5 performance changes.
A second change that is possible is uh
to rearchitect the way you are using
PHP. You can for example use franken PHP
or any other application server in PHP
to um save time bootstrapping PHP and
reuse state across requests. You can um
move processing out to the background
jobs using message cues and other um
patterns that allow you to reduce the
time spent in the web server. So what is
really possible to make a difference in
application performance is in your
hands. You are um driving this change by
building your application in a way that
ends up being faster for users. If you
like my content on PHP performance, you
can uh continue with this uh video on
PHP 8.5 changes in performance or you
can follow this channel or subscribe to
our newsletter link is in the
description. Bye.