INTRODUCING CLOUDBLAST WORKERS
Git Push to Live URL | Per Minute Metering | 100+ Templates | Salt Lake City / Hong Kong (UK Coming Soon)
Hello LowEndTalk Community!
You already know us for hourly billed AMD EPYC VPS. Today we are launching something new: Workers, our app platform.
Think Railway, Render, Fly.io or Vercel, but priced the way LET expects things to be priced. Point us at a Git repository, a Docker image or one of our templates, pick a datacenter, and your app is online with a public URL and SSL. No Dockerfile required, no pipeline to write, no server for you to patch or babysit.
Deploy now: https://console.cloudblast.io/
Product page: https://cloudblast.io/workers
LAUNCH GIVEAWAY: 10€ FREE CREDIT
To celebrate the launch we are giving 10€ of account credit to one lucky LET member.
How to enter
Just comment:
BLAST MEOne winner is picked at random and announced 30 days after this thread was posted. The credit works on Workers, on our VPS, or on both.
WHAT IS A WORKER?
A worker is an application, service or database that we build and run for you in an isolated container. You give it a repo, an image or a template, choose the region and the resources it may use, and it goes live. You never touch a server.
Any language, any framework
Node, Python, Go, Rust, PHP, Ruby and more, or bring your own Dockerfile.
Zero configuration
We detect the build, install the dependencies and start the process for you.
One click back to a working version
Every deploy is kept, so a bad release is undone in seconds.
Persistent disks
Attach a disk to any worker and it survives redeploys, restarts and scaling. Postgres, MySQL, Redis and MongoDB all work.
Logs, metrics and SSL included
Nothing extra to bolt on, nothing extra to pay for.
HOW IT WORKS
- Connect a Git repository, a Docker image or one of the 100+ templates.
- Build happens automatically on every push, with instant rollback to any earlier version.
3. Run with a public URL, SSL, logs, metrics and a persistent disk if it needs one.
PRICING: FOUR METERS, NOTHING ELSE
No plans to choose, no seats to buy, no minimum to clear. Usage is measured per minute and only while the worker runs, then charged at the hourly rates below. Stop a worker and the CPU and RAM meters stop with it. Run something for 8 minutes and you pay for 8 minutes, not for a full hour.
| Meter | Price | Unit |
|---|---|---|
| CPU | €0.011 | per core hour |
| RAM | €0.004 | per GB hour |
| Disk | €0.0001 | per GB hour |
| Traffic | €0.01 | per GB out |
Measured per minute, charged at the hourly rates above, invoiced once a month. 1 vCPU + 2 GB RAM running the full month lands at roughly 14€.
THE SAME WORKLOAD, A FRACTION OF THE BILL
One vCPU and 2 GB of RAM, for the hundred hours a month that a scheduled job or a staging environment actually runs:
| CloudBlast | Fly.io | Railway | Vercel | |
|---|---|---|---|---|
| vCPU, per hour | €0.0110 | n/a | €0.0252 | €0.1178 |
| RAM, per GB hour | €0.0040 | €0.0063 | €0.0126 | €0.0098 |
| Outgoing traffic, per GB | €0.01 | €0.02 | €0.05 | €0.14 |
| Minimum monthly fee | none | none | €4.60 | €18.40 |
| 1 vCPU + 2 GB, 100 hrs/month | €1.90 | €4.06 | €5.04 | €18.40 |
| 2.1x | 2.7x | 9.7x |
Public list prices taken from each provider's own pricing page in September 2026 and converted from US dollars at $1 = €0.92. The monthly figure is what the workload actually costs: the cheapest plan that allows it, plus any usage the plan's included credit does not cover. "n/a" means the resource is not sold by the unit, only inside a fixed machine size or plan.
100+ TEMPLATES, ONE CLICK AWAY
Frameworks, databases, queues, dashboards and automation tools, already configured. Pick one, name it, deploy.
Frameworks: Next.js, Nuxt, Node.js, Express, NestJS, Astro, Remix, SvelteKit, Django, FastAPI, Flask, Laravel, Ruby on Rails, Spring Boot, Go, Rust, Deno, Bun, PHP, Python
Data: PostgreSQL, MySQL, MariaDB, MongoDB, Redis, ClickHouse, Elasticsearch, Meilisearch, MinIO, RabbitMQ, Supabase, PocketBase
Apps and tooling: WordPress, Ghost, Strapi, Directus, n8n, Grafana, Prometheus, Metabase, Plausible, Umami, Gitea, Mattermost, Nextcloud, Ollama, Discord and Telegram bots
...and 250+ more.
WHERE WORKERS RUN
Two regions at launch, one more on the way. You pick the region per worker.
| Location | Status | Looking glass IPv4 | Looking glass IPv6 |
|---|---|---|---|
| Salt Lake City, USA | Live | 192.166.82.235 | 2a13:9500:3f:d7::0 |
| Hong Kong | Live | 178.83.121.94 | 2a0e:97c0:181:59::0 |
| Birmingham, UK | Coming soon | n/a | n/a |
Live latency tests and download files for each region: https://cloudblast.io/workers
MOVING OVER FROM RAILWAY, RENDER, FLY.IO OR VERCEL?
In most cases you will not need to change any code. Connect the same repository or Docker image, set the environment variables, attach a disk if you need one, and switch your domain over. If something does not fit, our support is available 24/7 and we will look at it with you.
PAYMENT METHODS
Cryptocurrencies (XMR included), credit cards, Alipay, WeChat Pay and 20+ local payment methods.
First time signing up, a 10€ deposit is required to start deploying. After that any amount works with usage based billing.
QUICK FAQ
Is there a plan, a seat fee or a minimum?
No. No plans, no per user seats, no minimum spend. You pay for the resources your workers use and nothing else.
Can I run a database?
Yes. Attach a persistent disk and it survives redeploys, restarts and scaling.
How granular is the billing?
Usage is measured per minute and charged at the hourly rates. A worker that ran for 20 minutes costs a third of an hour, not a full one. Invoices go out monthly.
What happens when I stop a worker?
The CPU and RAM meters stop with it. Disk keeps metering while the data is kept.
Do I need a Dockerfile?
No, but you can bring one if you prefer it.
DEPLOY YOUR FIRST WORKER: console.cloudblast.io
Questions, feature requests or a repo that will not build? Post here and we will answer, or join our Discord: https://discord.gg/7M84Xp8QBr
Happy deploying!