- It's designed for monorepos as you need to export/import the type definitions from/to your server
- If you're already in a team where you're mixing languages or have third party consumers that you have no control of, you're better off with making a GraphQL-API which is language-agnostic.
If you are already have a custom GraphQL-server for your project you may not want to use tRPC. GraphQL is amazing; it's great to be able to make a flexible API where each consumer can pick just the data needed for it.
The thing is, GraphQL isn't that easy to get right - ACL is needed to be solved on a per-type basis, complexity analysis, and performance are all non-trivial things.
We've taken a lot of inspiration from GraphQL & if you've made GraphQL-servers before you'll be familiar with the concept of input types and resolvers.
tRPC is a lot simpler and couples your server & website/app more tightly together (for good and for bad). It makes it easy to move quickly, do changes without updating a schema & there's no of thinking about the ever-traversable graph.
I've gotten asked several times about differences with Blitz.js and started outlining some differences on Twitter. If you think the below comparison is wrong in any way, please don't hesitate to reach out.
— Alex / KATT
The philosophy of the "Zero-API data layer" is the main common denominator in tRPC & Blitz but it is solved in different ways - Blitz solves the data layer by a custom compiler & tRPC solves the data layer by advanced type inference.
The key difference is that Blitz is a full-stack framework and tRPC is a library mainly targeting the data layer and doesn't have strong opinions about the rest of your stack.
- Easy HTTP caching of queries as requests are made with
- Query batching out-of-the-box
- Not tied to React as it has no custom build pipeline
- Works great with React Native
- Well-tested & production-ready
- Can be added to existing brownfield projects and adopted incrementally
- WebSockets transport
- Subscription support
- Zero-conf SSR with
_app.tsxdoes a prepass of all
useQueryon the server
- Configurable data flow between client/server with links
- Single API-endpoint which reduces amounts of cold starts in serverless environments
- More batteries included and easier to setup
- Blitz has a plenty of nice DX-benefits with it being a full-stack famework
- Blitz' custom build pipeline enables stuff like CMD+clicking a function to jump to the backend function
- Code scaffolding, plug-and-play auth, etc
- Blitzjs has a bigger community