Better Auth vs Clerk
Two sides of the authentication decision: library in your app and hosted service. When each fits, what it costs, who moves from one to the other, and what makers who chose it say.
Which fits you
- You want users in your own database and no per-user bill
Use it whenYou'd rather own the code than pay per user.
Trade-offYou host it and keep it patched; UI is yours to build.
- You're on Next.js or React and want sign-in, profiles and teams working today
Use it whenYou want prebuilt components, user profiles and organizations out of the box.
Trade-offPer-user pricing grows with you; users live in Clerk, not your database.
At a glance
| Used by | 6 makers' products · 94 open-source projects | 305 makers' products · 11 open-source projects |
|---|---|---|
| Cost at default usagemonthly active users 10k MAU | $0/mo Self-hosted (open source) | $0/mo Hobby |
| Moved to it on GitHubpull requests since Oct 2024 | 33 from Clerk | 4 from Better Auth |
| Downloads | 6.5M/wk8.4× vs npm | 2.1M/wk+50% vs npm |
| Pricing | Free (open source). | Free tier by monthly active users; paid plans add features and seats. · paid from $25/mo |
| Free tier | Yes | Yes |
| Open source | Yes · self-hostable | No |
| Incidents, 90 daysfrom its status page | no public status feed | 13 (1 major) |
Cost as you grow
Both cost $0 up to 50k MAU; from 100k MAU Better Auth costs less ($0 vs $1,025); and still does at 1M MAU ($0 vs $17,225). They're different kinds of tool — library in your app and hosted service — so the prices don't buy the same thing.
The numbers, plan by plan
| Monthly active users | Better Auth | Clerk |
|---|---|---|
| 100 | $0 Self-hosted (open source) | $0 Hobby |
| 1,000 | $0 Self-hosted (open source) | $0 Hobby |
| 5,000 | $0 Self-hosted (open source) | $0 Hobby |
| 10,000 | $0 Self-hosted (open source) | $0 Hobby |
| 25,000 | $0 Self-hosted (open source) | $0 Hobby |
| 50,000 | $0 Self-hosted (open source) | $0 Hobby |
| 100,000 | $0 Self-hosted (open source) | $1,025 Pro |
| 250,000 | $0 Self-hosted (open source) | $3,725 Pro |
| 1,000,000 | $0 Self-hosted (open source) | $17,225 Pro |
From each vendor's pricing page: Better Auth, Clerk.
Who moves from one to the other
Public pull requests on GitHub since Oct 2024 whose title says "Better Auth to Clerk" or the reverse — real code changes, by developers in general rather than makers only.
- feat: migrate from Clerk to Better AuthGoogle-Developer-Group-on-Campus-CTU/gdg-ctu-website · 2026-09-23
- feat(auth): Clerk to self-managed Better Auth cutovercoderclub-team/pixaPOS · 2026-09-18
- refactor: migrate authentication from Clerk to Better Authnomtan/cat-sns-app · 2026-09-03
- Migrate authentication from Clerk to Better AuthRookledger/GroundworkOS · 2026-08-28
- Migrate authentication from Clerk to Better AuthRookledger/GroundworkOS · 2026-08-28
- feat(auth): migrate Tinlance from Clerk to Better Auth + NeonLloydCoder/Tinlance · 2026-08-17
- feat: migrate from Clerk to self-hosted Better Auth (ADR-031)woosung-dev/kairos · 2026-08-16
- Migrate authentication from Clerk to Better AuthDawnMD/tutor-os · 2026-08-09
- P0: migrate iOS auth from Clerk SDK to Jovie Better AuthJovieInc/LogYourBody · 2026-07-25
- Migrate auth from Clerk to Better Auth (passkeys + email OTP)g30r93g/ID-Game · 2026-07-10
- fix: migrate auth from Better Auth to Clerkmarcojsmith/AgriBid · 2026-09-05
- Migrate authentication from Better Auth to Clerkrobertguss/web-app-starter-kit · 2026-08-05
- refactor: migrate authentication from Better Auth to Clerkbashirahmedchandio1/workos-ai · 2026-07-21
- Auth: remove Better Auth and cut over to Clerkprazgaitis/march-fit · 2026-02-25
What makers say
Makers on using it for authentication, from Product Hunt and Starter Story interviews, each linked to the source. Products with a page of their own and fuller notes first.
Authentication is always a time sink, but not with better_auth. Simple, secure, and extendable. Got magic links and login flow done in under an hour.
Better-Auth gave us a clean, flexible auth layer with minimal code. Performance, adaptability, and dev-experience were all wins.
It came out of nowhere and became popular for a good reason - it's really good!
The team loves Clerk for its intuitive interface, seamless integration, and efficient management of administrative tasks, allowing them to focus on building groundbreaking solutions.
We used Clerk for account management. Building an authentication infrastructure from scratch is challenging, but by using Clerk’s abstracted API, we were able to simplify this process.
Nobody wants to worry about authentication or build their own system. Clerk allowed me to simply integrate them into my project and immediately feel confident in the security of my site.
Loved and watch-outs
Themes that recur in makers' words and Hacker News comments, each linked to what it summarises.
- A library that keeps users and sessions in your own database, with no vendor dependency and no per-user fees. HNHN 2HN 3HN 4
- Plugins cover OAuth providers, magic links, RBAC, SSO and SAML, going well beyond Auth.js. HNHN 2HN 3PH
- Hackable and transparent, so custom flows such as iframe login work without fighting hidden internals. HNHN 2PH
- Built-in assumptions force hacks, such as OIDC providers without email or request-header-based admin scripts. HNHN 2
- Code quality felt rushed, with few tests and little logging, and audit logs need the managed service. HNHN 2
- Breaking changes appear in patch releases, and the OpenAPI spec gets little care for non-JS clients. HN
- Sign-up, sign-in, sessions and social login get wired in an afternoon instead of a sprint. PHPH 2
- Organizations, roles and billing come built in, covering multi-tenant B2B apps without custom identity work. PHPH 2
- Prebuilt UI components, including profile and settings pages, look polished out of the box. PH

