Push your code. RaidFrame builds, deploys, and auto-scales across 14+ global regions. From zero to millions of requests — no infrastructure to manage.
Three commands. No Dockerfile required. No infrastructure to configure.
Connect your GitHub repo or run rf deploy from the CLI. RaidFrame auto-detects your stack — Node.js, Python, Go, Ruby, Rust, Java, or any Dockerfile.
Docker images are built with layer caching (10s rebuilds). Rolling deployments ensure zero downtime. Health checks gate traffic to new instances.
CPU, memory, or request-based scaling. New instances launch in under 10 seconds. Scale to zero when idle, burst to hundreds during traffic spikes.
Bring a Dockerfile or let us auto-detect. If it runs in a container, it runs on RaidFrame.
Start free. Scale to dedicated 32-vCPU machines. Flat per-service pricing — no per-request charges, no bot traffic bills.
| Plan | CPU | RAM | Storage | Price |
|---|---|---|---|---|
| StarterFree tier | Shared | 512 MB | 1 GB | Free |
| Standard | 1 vCPU | 1 GB | 5 GB | $7/mo |
| Pro | 2 vCPU | 4 GB | 20 GB | $25/mo |
| Pro Plus | 4 vCPU | 8 GB | 40 GB | $50/mo |
| Performance | 8 vCPU | 16 GB | 80 GB | $100/mo |
| Performance XL | 16 vCPU | 32 GB | 160 GB | $200/mo |
| Dedicated | 32 vCPU | 64 GB | 320 GB | $400/mo |
All plans include: auto-scaling, zero-downtime deploys, SSL, private networking, and logs. No hidden fees.
Production infrastructure out of the box. No plugins, no add-ons, no third-party integrations.
Scale from 0 to hundreds of instances on CPU, memory, requests, or custom metrics. New instances healthy in under 10 seconds. Configurable cooldown prevents flapping.
Rolling deployments with health check gating. New instances must pass health checks before receiving traffic. Old instances drain gracefully. Every deploy is reversible.
Deploy to 14+ regions worldwide. Automatic failover routes traffic to the nearest healthy region. Add a region with one command: rf regions add eu-west-1.
Bring a Dockerfile or let RaidFrame auto-detect your stack. Layer caching means rebuilds in seconds. Multi-stage builds supported for optimized images.
Every pull request gets a full isolated environment with its own URL and database branch. Reviewers test real deployments, not code diffs. Auto-cleanup on merge.
All services communicate over encrypted private networks. Internal traffic never touches the public internet. Reference other services by name: http://api.internal:8080.
Drop into a running container with rf ssh. Run one-off commands with rf exec. Port-forward to your local machine for remote debugging with VS Code or Chrome DevTools.
Idle services scale to zero to save costs. First request triggers a cold start in under 3 seconds. Configure idle timeout per service. Only pay when your app is actually running.
Run migrations, seed caches, or warm up data before your main process starts. Init containers run sequentially — if any fails, the deployment rolls back automatically.
| Feature | AWS ECS | Vercel | Railway | RaidFrame |
|---|---|---|---|---|
| Deploy time | Minutes | ~60s | ~60s | ~30s |
| Auto-scaling | Config heavy | Serverless only | Manual | Built-in |
| Pricing model | Usage-based | Per-seat + usage | Usage-based | Flat per-service |
| Managed databases | RDS (separate) | Third-party | Built-in | Built-in |
| SSH access | Via SSM | No | No | Yes |
| Preview environments | Manual | Yes | No | Yes + DB branch |
| Background jobs | Fargate tasks | No | Separate service | Native |
| Setup complexity | Days | Minutes | Minutes | Minutes |
New instances spin up in under 10 seconds. Scale-down uses a configurable cooldown (default 5 minutes) to prevent flapping.
Yes. Bring a Dockerfile or let RaidFrame auto-detect your stack with buildpacks. Node, Python, Go, Ruby, Rust, Java, PHP, .NET — all supported.
Rolling deployments: new instances are health-checked, then traffic shifts gradually. Old instances drain connections. Zero downtime, every time.
Yes. The Starter plan includes a shared-CPU instance with 512 MB RAM. No credit card required. No time limit.
Yes. Deploy to 14+ regions with rf regions add. Traffic is routed to the nearest healthy region automatically.
Flat per-service pricing. No per-request charges, no bandwidth overage, no per-seat fees. Bot traffic doesn't affect your bill.
Free tier available. No credit card. No infrastructure setup. Just push and go.