A managed Postgres built like a precision tool: branches in a second, compute that disappears, recovery to the exact instant, and everything driven by API. Nothing invented — this is what the platform actually does.
No credit card — 3 projects included.
See also pricing and the API reference.
Postgres compute falls asleep on its own after idling and wakes on the first packet. You only pay for the seconds that are actually active.
-- compute suspended, resumed in under 1s $ psql "postgresql://…@db.lampion.cloud/ep-abc.postgres" SSL connection (protocol: TLSv1.3) -- you pay nothing while suspended mydb=> SELECT now(); 2026-08-04 10:42:01.234+00
Copy-on-write: a branch only stores its delta. Create one to test a migration, fill it with fixtures, throw it away afterwards.
# copy-on-write fork, isolated, in ~0.4s $ lampion branches create --from main --name staging → br-staging · ep-staging · 0.4s # delete it when you are done $ lampion branches delete staging
A full SQL editor, real-time observability and regression detection — all without ever leaving the browser.
-- Explain Analyze in one click from the editor mydb=> EXPLAIN ANALYZE SELECT * FROM users WHERE email ILIKE '%@corp.com'; Seq Scan on users Filter: (email ~~* '%@corp.com'::text) Planning Time: 0.08 ms Execution Time: 0.03 ms
Point-in-time recovery by LSN, restore to a fresh branch without touching production, and full portability between clouds.
# capture a restore point (LSN) $ lampion backups create --branch main --name before-migration → bk-42 · lsn 0/15A2C08 # restore into a fresh branch $ lampion backups restore bk-42 → branch restored-bk-42 · ep-new
Cloud-to-cloud migration has its own page — see how it works
Every connection is authenticated and encrypted, every project gets its own secrets, and your data is encrypted on every layer.
# TLS 1.3 enforced on every connection $ psql "postgresql://…@db.lampion.cloud/…?sslmode=require" SSL connection (TLSv1.3, cipher: TLS_AES_256_GCM_SHA384) # API: Bearer token scoped per organization $ curl -H "Authorization: Bearer lmp_live_…" \ api.lampion.cloud/v1/projects
Today, the whole data plane (compute, K8s, S3) runs on Scaleway, in Paris — a single console. No lock-in: the infrastructure is described as multi-provider Terraform, and portability to a second sovereign French cloud is on the roadmap.
| Layer | Operator | Site | Attestation |
|---|---|---|---|
| compute | Scaleway | Paris — fr-par-1 | ISO/IEC 27001 · HDS |
| pageserver · WAL | Scaleway | Vitry-sur-Seine | ISO/IEC 27001 · HDS |
| S3 storage (backups) | Scaleway | fr-par — Vitry-sur-Seine | ISO/IEC 27001 · HDS · AES-256 |
| control plane | Lampion SAS | Paris, France | RCS Paris |
Invite by email, assign granular roles, and let everyone switch between organizations instantly — every action is logged.
# role assigned at invitation time $ lampion members invite [email protected] --role developer → inv-9a · developer · pending
A public REST API, signed webhooks and transparent billing — all embeddable in your CI/CD pipelines, your CLI or your agents.
$ curl -X POST \ -H "Authorization: Bearer lmp_live_…" \ -d '{"name":"my-app","region":"fr-par-1"}' \ api.lampion.cloud/v1/projects # project + connection string, ready to use {"id":"abc123","endpoint_id":"ep-abc", "connection_string":"postgresql://… @db.lampion.cloud/ep-abc.postgres?sslmode=require"}
Full reference for the 40+ endpoints. The full plan and quota breakdown is on the pricing page.