[tissue](/) · [comparisons](/vs/) · tissue vs supabase

# Tissue vs Supabase

Supabase is the back of the house. Tissue is the back of the house and the front. That is most of the difference; the rest is Postgres against SQLite.

If your application is fundamentally a Postgres application — extensions, row-level security, realtime subscriptions, a table editor your colleagues actually open — Supabase is built around exactly that and we are not.

Two things bring people across. Supabase does not host your website, so there is a second deploy and a second bill somewhere else; and a free project is paused after a week of inactivity, with a limit of two active ones. On tissue the site, the code and the data are one deploy, and a quiet project stays up.

## What each one is

Two sentences each, before any numbers.

- **Supabase** — A hosted Postgres database with a generated REST and realtime API over it, plus authentication, file storage and edge functions. You get a project, a dashboard and a connection string; the website that uses them lives on another host.
- **Tissue** — One Cell — a JavaScript or WebAssembly function answering on a URL — with a SQL database, object storage, scheduled runs, sign-in and secrets attached to it by name in `ribo.toml`. `ribo deploy` puts the code, the static files and every attachment live in one step.

## Side by side

Every figure below came off the other company’s own pages, on one day, and is dated underneath.

|   | Supabase | Tissue |
| --- | --- | --- |
| Free plan | $0. 500 MB of database on shared CPU, 1 GB of file storage, 5 GB of egress, 50,000 monthly active users, 500,000 function invocations. Projects are paused after a week of inactivity, and you may have two active at once. | $0, no card. 5 Cells, 1 GiB of database, 5 GiB of file storage, 2,000,000 CPU-milliseconds a month. Nothing pauses, and there is no active-project limit. |
| Entry paid plan | Pro, $25 a month, including $10 of compute credit — enough for one Micro instance. The next size up is $15 a month on top of the plan. | Strand, $5 a month. There is no instance under it. |
| What is metered | Database size, file storage, egress, monthly active users, function invocations, realtime messages, and the compute instance the project runs on. | CPU-milliseconds and gigabytes stored. |
| Database | Postgres. 8 GB of disk included on Pro, then $0.125 per GB. | c3, SQL over SQLite. 5 GiB on Strand, then $0.50 per GiB-month. |
| Vector search | pgvector, an extension you enable on the project. | Built into c3: an `F32_BLOB` column, cosine or L2 distance, and an approximate-nearest-neighbour index. |
| File storage | 100 GB on Pro, then $0.0213 per GB. | g7. 20 GiB on Strand, then $0.015 per GiB-month. |
| Egress | 250 GB on Pro, then $0.09 per GB. | Free and not metered. |
| Sign-in for your users | A full authentication product: many providers, 100,000 monthly active users on Pro, then $0.00325 each. | gate: an email domain or a list of addresses. Nobody is counted. |
| Hosting the website | Not included. The site goes on another host, with its own deploy and its own bill. | Included. Everything under `public/` is served at the root of the same deploy. |
| Scheduled jobs | Through a Postgres extension, inside the database. | pulse, a cron line in the config. 3 schedules free, 10 on Strand. |

Checked 16 September 2026 against Supabase's own pricing page. Both of us move; if a figure here has gone stale, [tell us](mailto:hello@tissue.systems?subject=vs/supabase) and we will correct it.

## Where Supabase is the better choice

This section is not decorative.

Postgres is the reason. Extensions, JSON that behaves, row-level security as a policy language, listen/notify, and a planner that has seen everything. c3 is SQLite: small, fast, enough for most sites, and not Postgres.

Their authentication is a product; ours is a door. Supabase does social providers, magic links, multi-factor and per-row authorization tied to the signed-in user. gate decides whether somebody gets in at all. If you need accounts with profiles and permissions, you would be writing that yourself here.

Realtime. Subscribing a browser to a table's changes is one line there. Cells hold no long-lived connections, so that pattern has no equivalent here.

The dashboard. A table editor, a SQL editor, logs and a schema view that somebody who does not deploy can be pointed at.

## Where tissue is the better choice

The site and the database in one deploy. No second host, no CORS between your front end and your data, no second renewal date.

A quiet project stays up. Nothing is paused after a week, and there is no ceiling of two.

No instance underneath. Supabase's paid price starts with a machine you rent whether or not anybody visits; ours starts at the CPU you actually spent.

Egress is free. 250 GB is plenty until the week it is not, and after that it is nine cents a gigabyte.

## What moving looks like

The front end and the queries move into the same Cell: the pages come across as static files, and the calls that went over the network to a project somewhere else become `env.DB`.

```toml
[cell]
name = "app"
js   = "./cell.js"

[[bindings]]
type    = "files"
binding = "FILES"
dir     = "public"

[[bindings]]
type     = "c3"
binding  = "DB"
database = "app"
```

```
$ ribo deploy
uploading   cell.js
publishing  ./public
attaching   c3

live  https://app.quartz-4b.tissue.dev
```

The database and the bucket are created on that first deploy if they are not there yet. `ribo deploy` is the whole pipeline: there is no build step to configure and no second place to put the static files.

Spore is $0, takes no card, and allows commercial use. If tissue turns out to be the wrong shape for what you are building, you will know inside an afternoon.

[Start free](https://auth.tissue.systems/login?plan=spore#signup) 5 Cells, a SQL database, file storage and cron.

## The other comparisons

- [vs Vercel](/vs/vercel)
- [vs Cloudflare](/vs/cloudflare)
- [vs Val Town](/vs/val-town)
- [vs Render](/vs/render)
- [vs Fly.io](/vs/fly-io)
- [vs Squarespace](/vs/squarespace)