diff --git a/apps/blog/content/blog/prisma-next-performance-benchmark/index.mdx b/apps/blog/content/blog/prisma-next-performance-benchmark/index.mdx index f19b8fa593..66d2ee9bad 100644 --- a/apps/blog/content/blog/prisma-next-performance-benchmark/index.mdx +++ b/apps/blog/content/blog/prisma-next-performance-benchmark/index.mdx @@ -1,11 +1,11 @@ --- -title: "Prisma Next is 50% faster than Prisma 7: a performance deep dive" +title: "Prisma Next Is ~90% As Fast as Raw PG" slug: "prisma-next-performance-benchmark" date: "2026-06-11" authors: - "Serhii Tatarintsev" -metaTitle: "Prisma Next is 50% faster than Prisma 7 | Benchmark" -metaDescription: "Prisma Next reaches about 50% higher peak throughput than Prisma 7 and holds latency low under load, closing most of the gap to the raw pg driver while keeping Prisma's type-safe, model-first workflow." +metaTitle: "Prisma Next Is ~90% As Fast as Raw PG" +metaDescription: "Prisma Next performance benchmarks achieves ~90% of the raw pg driver's peak throughput, holds latency low under load, and ships as a 148.5 KB gzipped bundle for serverless and edge workloads." heroImagePath: "/prisma-next-performance-benchmark/imgs/hero.svg" heroImageAlt: "Bar chart showing Prisma Next outperforming Prisma 7 in throughput benchmarks" metaImagePath: "/prisma-next-performance-benchmark/imgs/meta.png" @@ -26,7 +26,7 @@ To see where it stands, we ran the same Postgres workload through three setups: - At 6,000 to 7,000 requests/second, a level both can serve, Prisma Next's p95 latency stays around 4 ms while Prisma 7's has already climbed past 40 ms. - Prisma Next reaches about **87%** of the raw `pg` driver's peak, while Prisma 7 reaches about **58%**. -It's also about **9× smaller** to ship, which matters for serverless and edge deployments where bundle size drives cold starts. We look at that below. +Its bundle size gzipped is about **148.5 KB** gzipped, compared to **1.32 MB** for Prisma 7, also about **9× smaller** to ship, which matters for serverless and edge deployments where bundle size drives cold starts. We look at that below. One note before the numbers: **Prisma 7 is still the version to use in production today.** [Prisma Next is in Early Access](https://pris.ly/pn-ea), and once it's Generally Available you'll be able to run it in production too. @@ -44,7 +44,7 @@ The raw `pg` driver is our baseline: the fastest this machine can talk to the da This is the main result. As traffic rises, all three lines climb together at first; then Prisma 7 flattens out, while Prisma Next and `pg` keep climbing for thousands more requests per second. -![Throughput over the load test: Prisma 7 flattens out just under 8,000 req/s while Prisma Next and pg keep climbing](/prisma-next-performance-benchmark/imgs/throughput-ramp.svg) +![Throughput over the load test: Prisma 7 flattens out just under 8,000 req/s while Prisma Next and pg keep climbing](/prisma-next-performance-benchmark/imgs/throughput-ramp.png) That flat line is the overhead the new foundation was built to reduce. The peak each setup sustains, measured as the average of its busiest 10% of the run, lands here: @@ -95,4 +95,20 @@ A few caveats. This is one benchmark, on one workload and one machine, so result We've shared [our benchmark repo](https://pris.ly/pn-benchmarks), so you can run it yourself and tell us where we're wrong. We'll keep testing more workloads and tuning as Prisma Next moves toward Prisma 8. -To try Prisma Next today, start with the [Early Access blog post](https://pris.ly/pn-ea). Star and watch [prisma/prisma-next](https://github.com/prisma/prisma-next) on GitHub to follow what ships next, and if you hit a snag, start a thread in [#prisma-next on our Discord](https://pris.ly/discord). +## Try Prisma Next today + +There are two ways to get started. To spin up a complete template app, scaffold one with: + +```shell +bunx create-prisma@next +``` + +This sets you up with a [Prisma Postgres](https://www.prisma.io/postgres) database and can deploy the app to [Prisma Compute](https://www.prisma.io/compute), our serverless runtime for Prisma apps, so you can see the bundle-size and cold-start benefits above end to end. + +To add Prisma Next to an existing app, run: + +```shell +bunx prisma-next@latest init +``` + +To learn more, start with the [Early Access blog post](https://pris.ly/pn-ea). Star and watch [prisma/prisma-next](https://github.com/prisma/prisma-next) on GitHub to follow what ships next, and if you hit a snag, start a thread in [`#prisma-next` on our Discord](https://pris.ly/discord). diff --git a/apps/blog/public/prisma-next-performance-benchmark/imgs/hero.svg b/apps/blog/public/prisma-next-performance-benchmark/imgs/hero.svg index 1e6f4b3423..9f5258df5e 100644 --- a/apps/blog/public/prisma-next-performance-benchmark/imgs/hero.svg +++ b/apps/blog/public/prisma-next-performance-benchmark/imgs/hero.svg @@ -1,22 +1,22 @@
Throughput benchmark
Prisma Next
Performance
Requests / second
+50%faster
Prisma Next
12,534req/s
Prisma 7
8,260req/s
 \ No newline at end of file +}]]>
Throughput vs raw pg
Prisma Next Is
~90% As Fast
as Raw PG
pg (raw driver)
14,350req/sec
Prisma Next
12,534req/sec
\ No newline at end of file diff --git a/apps/blog/public/prisma-next-performance-benchmark/imgs/meta.png b/apps/blog/public/prisma-next-performance-benchmark/imgs/meta.png index 673f1a56fd..43ef6f7cb8 100644 Binary files a/apps/blog/public/prisma-next-performance-benchmark/imgs/meta.png and b/apps/blog/public/prisma-next-performance-benchmark/imgs/meta.png differ diff --git a/apps/blog/public/prisma-next-performance-benchmark/imgs/throughput-ramp.png b/apps/blog/public/prisma-next-performance-benchmark/imgs/throughput-ramp.png new file mode 100644 index 0000000000..fdd0dafbf5 Binary files /dev/null and b/apps/blog/public/prisma-next-performance-benchmark/imgs/throughput-ramp.png differ diff --git a/apps/blog/public/prisma-next-performance-benchmark/imgs/throughput-ramp.svg b/apps/blog/public/prisma-next-performance-benchmark/imgs/throughput-ramp.svg deleted file mode 100644 index ac13c6dd87..0000000000 --- a/apps/blog/public/prisma-next-performance-benchmark/imgs/throughput-ramp.svg +++ /dev/null @@ -1 +0,0 @@ -Throughput climbs until each setup hits its ceilingRequests per second as load ramps up — Prisma 7 plateaus while Prisma Next keeps scalingPrisma NextPrisma 7pg (raw driver)03k6k9k12k15k0s60s120s180s240s300sTime into load testPrisma 7 ceiling ~8,300 req/s