---
title: "Discussion on \"React SSR Framework Showdown: TanStack Start, React Router, and Next.js Under Load\" | Hashnode"
description: "Discussion on \"React SSR Framework Showdown: TanStack Start, React Router, and Next.js Under Load\". We benchmarked TanStack Start, React Router, and Next.js running an identical SSR eCommerce app at 1,000 requests per second on Kubernetes."
url: "https://hashnode.com/posts/react-ssr-framework-benchmark-tanstack-start-react-router-nextjs/69b965b5760a8588a4c92835"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [React SSR Framework Showdown: TanStack Start, React Router, and Next.js Under Load](https://blog.platformatic.dev/react-ssr-framework-benchmark-tanstack-start-react-router-nextjs?utm_source=hashnode&utm_medium=feed)

We benchmarked TanStack Start, React Router, and Next.js running an identical SSR eCommerce app at 1,000 requests per second on Kubernetes.

Mar 17 · 14 min read · blog.platformatic.dev  
Updated Mar 24  
Read the article

-   [#nextjs](https://hashnode.com/tag/nextjs)
-   [#react-router](https://hashnode.com/tag/react-router)
-   [#nodejs](https://hashnode.com/tag/nodejs)
-   [#kubernetes](https://hashnode.com/tag/kubernetes)

## Written by

[Matteo Collina](https://hashnode.com/@matteocollina)

@matteocollina

  
[View profile](https://hashnode.com/@matteocollina)

Open Sourcerer, Node.js TSC member, Fastify Lead Maintainer

## Responses 3

Markdown supported · ⌘↵ to post

[Apurv Julaniya](https://hashnode.com/@apurvjulaniya)

[Mar 18](https://hashnode.com/posts/react-ssr-framework-benchmark-tanstack-start-react-router-nextjs/69b965b5760a8588a4c92835/comment/69baea558c55d6eefb48c72f)

This benchmark is interesting because it highlights something many developers miss: performance differences between frameworks are often more about architecture choices and defaults than raw capability. For example, TanStack Start’s explicit control and Vite-based pipeline can improve developer speed, while Next.js focuses more on server-first optimizations and ecosystem integration. � TanStack One thing I’d add is that real-world performance depends heavily on caching strategy, hydration cost, and app interactivity level, not just SSR benchmarks. A highly interactive app may behave very differently compared to a content-heavy site. It would be valuable to also compare Time to Interactive (TTI) and client-side navigation performance, since those often impact user experience more than initial SSR numbers. I’ve been exploring similar ideas around React architecture and performance trade-offs—this kind of data-driven comparison is exactly what the ecosystem needs.

[Danilo Velasquez Urrutia](https://hashnode.com/@dvelasquez)

[Mar 23](https://hashnode.com/posts/react-ssr-framework-benchmark-tanstack-start-react-router-nextjs/69b965b5760a8588a4c92835/comment/69c142b030a9b81e3a53c372)

How many processes you were running at the same time? For me the numbers don't add up.

[CapeStart](https://hashnode.com/@capestart)

[Mar 19](https://hashnode.com/posts/react-ssr-framework-benchmark-tanstack-start-react-router-nextjs/69b965b5760a8588a4c92835/comment/69bbf91a8c55d6eefbedfddc)

Really interesting benchmark. What stood out is how much framework performance depends on architecture choices and how quickly things can change when teams dig into the bottlenecks.
