Compare, hosted
Levelrail vs Cloud Run
A good fit for bursty traffic, with scale to zero and per-request billing.
As of 2026-10. Statements about Cloud Run come from its own public documentation, linked on each row.
The short verdict
Cloud Run is good at: scale to zero and per-request billing are hard to beat for spiky or rare traffic; and capacity can appear in seconds under load and span zones or regions; and google identity and networking features such as workload identity.
Choose Cloud Run when traffic is spiky or rare, you need capacity in seconds, or you depend on Google identity and networking features.
Levelrail differs here: a fixed-price server you run, so cost does not move with traffic; and replicas are fixed with no autoscaling, and wake from sleep is slower than a cold start; and logs, metrics and data stay on machines you control, with nothing billed by volume. Levelrail is pre-release, a single node is the well-tested path and multi-node is in beta, so its track record is much shorter than Cloud Run's.
Where Cloud Run is ahead
- Git push deploys from GitHub, GitLab, Bitbucket. Cloud Run: Developer Connect covers GitHub, GitLab, Bitbucket; Cloud Build covers GitHub. Levelrail: partly, all four documented; GitLab and Bitbucket previews are beta, synthetic-webhook tested.
- Builds from source without a Dockerfile. Cloud Run: Google Cloud buildpacks detect the language; no Dockerfile needed. Levelrail: partly, railpack detects Node, Python, Go and more; Ruby is rejected.
- Load balancing across replicas. Cloud Run: Requests balanced across regional instances automatically; multi region needs load balancer. Levelrail: partly, experimental and flag gated; not tried across real hosts.
- Autoscaling. Cloud Run: Scales instances on request volume, with configurable maximums. Levelrail: partly, beta: CPU, memory or request-rate scaling with cooldowns and reasons.
Where Levelrail is ahead
- Crashloop diagnosis with last log lines. Levelrail: Crashloop detection surfaces the last log lines. Cloud Run: partly, troubleshooting guide; logs checked manually, no automatic crashloop view.
- Readiness-gated deploys (zero-downtime on one node). Levelrail: Cutover only after the readiness probe passes. Live-tested on one node; multi-node is beta. Cloud Run: partly, gradual rollouts and traffic splits; health gating of cutover not stated.
- Preview environments per pull request. Levelrail: Stable on GitHub, beta on GitLab and Bitbucket. Cloud Run: partly, revision tags give separate URLs without traffic; no PR previews.
- Scheduled jobs. Levelrail: Per-app cron jobs with run history, logs, retries and alerts. Cloud Run: partly, scheduled job runs need Cloud Scheduler calling the job endpoint.
- Request firewall, rate limits, bot protection. Levelrail: Ordered rules, IP bans, bot handling, challenges and rate limits. Cloud Run: partly, cloud Armor filters traffic only through an external load balancer.
- MCP / AI assistant integration. Levelrail: MCP server with over 240 tools in on-demand toolsets. Cloud Run: partly, managed remote MCP server deploys and inspects services.
- IAM, audit log and 2FA in the free edition. Levelrail: Allow/Deny policies, audit log and TOTP in one edition. Cloud Run: partly, cloud IAM and audit logs; 2FA is not a Cloud Run feature.
28 capabilities side by side
Yes, partly and no are what each project's own documentation states. "Not stated" means their docs do not say, which is not the same as no. Levelrail maturity tags (beta, stable) come from its feature status page.
| Capability | Levelrail | Cloud Run |
|---|---|---|
| Architecture | ||
| No SSH: agents dial out | Yesbeta Agents dial out over mTLS; no inbound portsSource for Levelrail, No SSH: agents dial out | Not applicable |
| Level-triggered reconciler with status reasons | Yes Desired state converged, status condition with reason per passSource for Levelrail, Level-triggered reconciler with status reasons | Not applicable |
| Single control plane binary | Yes One control plane binary, plus one agent per extra nodeSource for Levelrail, Single control plane binary | Not applicable |
| Observability | ||
| Metrics and log search built in | Yes Node-local metrics at 15s and full-text log searchSource for Levelrail, Metrics and log search built in | Yes Automatic Cloud Monitoring metrics; logs in Cloud LoggingSource for Cloud Run, Metrics and log search built in |
| Crashloop diagnosis with last log lines | Yes Crashloop detection surfaces the last log linesSource for Levelrail, Crashloop diagnosis with last log lines | Partly Troubleshooting guide; logs checked manually, no automatic crashloop viewSource for Cloud Run, Crashloop diagnosis with last log lines |
| Deploys | ||
| One-click rollback with pinned images | Yes Prior images pinned so cleanup cannot remove themSource for Levelrail, One-click rollback with pinned images | Yes Traffic rolls back to a previous revision immediatelySource for Cloud Run, One-click rollback with pinned images |
| Readiness-gated deploys (zero-downtime on one node) | Yes Cutover only after the readiness probe passes. Live-tested on one node; multi-node is betaSource for Levelrail, Readiness-gated deploys (zero-downtime on one node) | Partly Gradual rollouts and traffic splits; health gating of cutover not statedSource for Cloud Run, Readiness-gated deploys (zero-downtime on one node) |
| Preview environments per pull request | Yesstable Stable on GitHub, beta on GitLab and BitbucketSource for Levelrail, Preview environments per pull request | Partly Revision tags give separate URLs without traffic; no PR previewsSource for Cloud Run, Preview environments per pull request |
| Docker Compose apps | Partly Pre-built images from the dashboard; git deploys can buildSource for Levelrail, Docker Compose apps | Partly Compose file deploys as one multi container service; alpha or previewSource for Cloud Run, Docker Compose apps |
| Git push deploys from GitHub, GitLab, Bitbucket | Partly All four documented; GitLab and Bitbucket previews are beta, synthetic-webhook testedSource for Levelrail, Git push deploys from GitHub, GitLab, Bitbucket | Yes Developer Connect covers GitHub, GitLab, Bitbucket; Cloud Build covers GitHubSource for Cloud Run, Git push deploys from GitHub, GitLab, Bitbucket |
| Builds from source without a Dockerfile | Partly Railpack detects Node, Python, Go and more; Ruby is rejectedSource for Levelrail, Builds from source without a Dockerfile | Yes Google Cloud buildpacks detect the language; no Dockerfile neededSource for Cloud Run, Builds from source without a Dockerfile |
| Scheduled jobs | Yes Per-app cron jobs with run history, logs, retries and alertsSource for Levelrail, Scheduled jobs | Partly Scheduled job runs need Cloud Scheduler calling the job endpointSource for Cloud Run, Scheduled jobs |
| Data | ||
| Managed databases with scheduled backups | Partly Many engines, scheduled S3 backups; Postgres restores tested liveSource for Levelrail, Managed databases with scheduled backups | Not stated in their docs |
| Encrypted secrets for env vars | Yes Envelope-encrypted, masked in UI, injected at container createSource for Levelrail, Encrypted secrets for env vars | Yes Secret Manager values injected as env vars or mounted volumesSource for Cloud Run, Encrypted secrets for env vars |
| Networking | ||
| Private networking between apps or nodes | Partly Named networks and allow rules; cross-server mesh unverified on real hostsSource for Levelrail, Private networking between apps or nodes | Partly Direct VPC egress reaches private IPs; internal DNS not documentedSource for Cloud Run, Private networking between apps or nodes |
| Load balancing across replicas | Partly Experimental and flag gated; not tried across real hostsSource for Levelrail, Load balancing across replicas | Yes Requests balanced across regional instances automatically; multi region needs load balancerSource for Cloud Run, Load balancing across replicas |
| Request firewall, rate limits, bot protection | Yes Ordered rules, IP bans, bot handling, challenges and rate limitsSource for Levelrail, Request firewall, rate limits, bot protection | Partly Cloud Armor filters traffic only through an external load balancerSource for Cloud Run, Request firewall, rate limits, bot protection |
| Automatic TLS certificates | Partly Internal issuer by default; public ACME verified once liveSource for Levelrail, Automatic TLS certificates | Partly Managed certificates on custom domain mappings, which are in PreviewSource for Cloud Run, Automatic TLS certificates |
| Platform | ||
| Multi-node | Partlybeta Enrollment verified locally; WireGuard mesh still unverifiedSource for Levelrail, Multi-node | Not applicable |
| One-click service templates | Yesbeta About 360 templates; a sample is live-deployed in e2eSource for Levelrail, One-click service templates | Not stated in their docs |
| Autoscaling | Partly Beta: CPU, memory or request-rate scaling with cooldowns and reasonsSource for Levelrail, Autoscaling | Yes Scales instances on request volume, with configurable maximumsSource for Cloud Run, Autoscaling |
| CLI and HTTP API | Yes CLI and REST API documented; MCP server is betaSource for Levelrail, CLI and HTTP API | Yes gcloud CLI and Admin API manage services and jobsSource for Cloud Run, CLI and HTTP API |
| Deploy and alert notifications | Partly 18 channel kinds in beta, tested only against mocksSource for Levelrail, Deploy and alert notifications | Partly Cloud Monitoring alerts to email, Slack, webhooks, Pub/SubSource for Cloud Run, Deploy and alert notifications |
| MCP / AI assistant integration | Yesbeta MCP server with over 240 tools in on-demand toolsetsSource for Levelrail, MCP / AI assistant integration | Partly Managed remote MCP server deploys and inspects servicesSource for Cloud Run, MCP / AI assistant integration |
| Runs on servers you own | Yes Runs on Linux servers you provideSource for Levelrail, Runs on servers you own | Not applicable |
| Open source license | Yes Apache 2.0, one edition, no paid tierSource for Levelrail, Open source license | Not stated in their docs |
| Security | ||
| IAM, audit log and 2FA in the free edition | Yesstable Allow/Deny policies, audit log and TOTP in one editionSource for Levelrail, IAM, audit log and 2FA in the free edition | Partly Cloud IAM and audit logs; 2FA is not a Cloud Run featureSource for Cloud Run, IAM, audit log and 2FA in the free edition |
| SAML or SCIM single sign-on | Partly SAML and SCIM are experimental and off by defaultSource for Levelrail, SAML or SCIM single sign-on | Not stated in their docs |
All ten projects are on the full matrix. Product names and logos belong to their owners and are used for identification only.
Migrating from Cloud Run
No importer. Deploy the same container image; the runbook maps each Cloud Run concept and lists what has none.
Questions about Levelrail and Cloud Run
Is Levelrail a good alternative to Cloud Run?
A good fit for bursty traffic, with scale to zero and per-request billing. How Levelrail differs: A fixed-price server you run, so cost does not move with traffic. Replicas are fixed with no autoscaling, and wake from sleep is slower than a cold start. Logs, metrics and data stay on machines you control, with nothing billed by volume. Choose Cloud Run instead when traffic is spiky or rare, you need capacity in seconds, or you depend on Google identity and networking features. Levelrail is pre-release, so check the comparison for what it does not do yet.
When should I choose Cloud Run instead of Levelrail?
Choose Cloud Run when traffic is spiky or rare, you need capacity in seconds, or you depend on Google identity and networking features.
Can I move from Cloud Run to Levelrail?
No importer. Deploy the same container image; the runbook maps each Cloud Run concept and lists what has none.
Is Levelrail ready to replace Cloud Run in production?
Pre-release: no stable release; APIs, the app spec and the on-disk layout can change between betas. A single node is the well-tested path; multi-node is in beta and the WireGuard mesh is not battle-tested. Check the feature status page before moving anything you cannot afford to lose, and keep Cloud Run running alongside until you trust the result.