Deploy your APIs, WebApps, Microservices, AI Agents, and Workloads Lyfted to the cloud
in seconds.

Streamline your Docker deployments on shared or dedicated cloud compute, in the region you choose.

Deploy your code See the platform →
No upfront charges · Billed by the second

See it running. Not a mockup.

These are real screens from the DevLyft dashboard. Every build, deploy and node you create shows up here, with the CPU, memory and per-second cost you're actually using.

app.devlyft.io/dashboard
DevLyft dashboard overview showing projects, deploys, builds, environments and a usage trend
The overview: projects, deploys, builds and environments at a glance, with a live usage trend and a shared-vs-dedicated tenancy breakdown.
/dashboard · Deployments
DevLyft deployments table listing deploys with build, location, tenancy, replicas and autoscaling range
Deployments. Every deploy in one table — its build, location, shared or dedicated tenancy, replica count and autoscaling range.
/dashboard · Nodes
DevLyft nodes table showing live node health, location, architecture, CPU, memory and IP
Nodes. Live health for every node across your locations — architecture, capacity and status, shared or dedicated.

View the full product walkthrough →

From repo to running, in four floors.

Get from build to running infrastructure in minutes — no runbooks required.

01 Step 01

Create your build

Add a Docker image to your project, OR, build straight from your repository with GitHub Actions Coming soon arrives with the DevLyft CLI.

02 Step 02

Choose location & node type

Pick where it runs and whether it's isolated or shared.

03 Step 03

Deploy with one click

Watch logs stream live and get a working URL in seconds.

04 Step 04

Scale automatically

Traffic grows, nodes follow. No runbooks required.

See the full product walkthrough →

Size it. See the price.

Billed per second on the CPU and memory you request — no subscriptions, no reserved capacity, no platform fee.

Pick the CPU and memory your workload reserves — you're billed per second on exactly that.

1
2 GB

1

From 0.125 vCPU / 256 MB (≈$3/mo) up to 4 vCPU / 8 GB · 1–4 GB memory per vCPU · each instance billed per second.

Shared

$27.44/mo

≈ $0.038/hr · billed per second

1.00 vCPU · 2.0 GB · 1 instance

Priced per instance, per month. If your app scales out, you pay per instance — so 2 instances is 2× — charged per second, only for the time the extra instances run.

    $0.05/GB egress (1 GB = 10⁹ bytes).

    Prices in USD. Monthly figures assume 730 hours. Memory GB = 1024 MB (GiB).

    Start deploying

    No upfront charges · billed per second

    Pick your location

    Launch in the region closest to your users, with live latency estimates before you deploy.

    Shared or dedicated

    Start on shared compute, move a workload to its own isolated node when it needs one. Switch anytime.

    Logs and metrics included

    Live logs, CPU and memory from the first deploy.

    No platform fee

    You pay for the CPU and memory you reserve, per second. Stop a workload and its compute billing stops.

    Need volume discounts, SSO, or a dedicated SLA? Talk to sales →

    Everything ops normally babysits, automated.

    One platform: bring your code, pick your compute, let it run. Push a Docker image (GitHub Actions is coming soon), choose isolated or shared, let it scale.

    Explore the full platform →

    Bring what you already have

    Deployments currently start from your own Docker image: push an existing image into a project and DevLyft turns it into a running deployment. Building and deploying directly from your repository with GitHub Actions Coming soon will be available when the DevLyft CLI launches.

    Pick your location

    Launch in the location closest to your users. Live latency estimates, no guesswork.

    Isolated & shared nodes

    Dedicate isolated compute to workloads that need it, run everything else on cost-efficient shared pools. Switch anytime.

    Automated scaling

    Coming soon

    Nodes spin up and down with real traffic, in any location, around the clock. No dashboards to watch at 2am.

    Logs and metrics from the first deploy

    Every deployment streams its logs live and reports CPU and memory, so you can see what it is doing before anything breaks.

    Your domain, verified

    Attach a custom domain to a deployment and prove you own it. DevLyft handles the routing from there.

    Secrets per environment

    Store database URLs and API keys against a project and environment, and reference them from your deployments instead of committing them to a repo.

    Deployment infrastructure, without the pager.

    An AI agent can set up a server for you in an afternoon, and for a static site Cloudflare Pages is hard to beat. The cost shows up afterward, when the disk fills at 2am or you need to see what a deployment is doing, and the agent that wrote the setup isn't there. DevLyft is for the backend that has to keep running.

    DevLyft is one platform that handles location placement and isolation for you, billed to the second, with automatic scaling coming soon. You supply the artifact. We keep it running, wherever it needs to run, for exactly as long as it's needed.

    Put plainly: DevLyft is a Platform-as-a-Service for developers. Push a container (GitHub Actions support is coming soon with the DevLyft CLI) and it runs on shared or dedicated nodes in the locations you choose, billed by the second on the exact CPU and memory you request — no clusters to operate, and no platform fee.

    Built for
    • Solo developers shipping a side project on a Friday night
    • Small teams without a dedicated ops person
    • Startups scaling from zero users to real traffic
    • Agencies hosting projects for multiple clients
    • Open-source maintainers running demos & docs sites
    • Students & hobbyists learning to ship real infra

    Ready to Lyft your next deployment?

    Pick a location, ship your code, and let DevLyft handle the rest.

    Start deploying Talk to sales
    No upfront charges · Cancel anytime