Podflareprivate alpha

Simple, predictable pricing

Start free with a $200 credit. Scale up as you ship. Every tier includes the full SDK, all frameworks (Anthropic, OpenAI Agents, Vercel AI SDK), and every region we operate in.

Private alphapaid upgrades activated by email within 15–30 min — no card hold

Free

$0 / mo

Everything you need to ship a prototype.

  • ✓5 min idle timeout per sandbox
  • ✓30 min max lifetime per sandbox
  • ✓1 GB RAM • 2 vCPU • 4 GB disk per sandbox
  • ✓10 concurrent sandboxes
  • ✓All regions (EU, US-West)
  • ✓Vercel AI SDK / Anthropic / OpenAI Agents adapters
Sign up free →

Pro

most popular
$50 / mo

Production agent workloads with longer sessions.

  • ✓30 min idle timeout (6× Free)
  • ✓8 hr max lifetime (16× Free)
  • ✓4 GB RAM • 4 vCPU • 16 GB disk per sandbox
  • ✓50 concurrent sandboxes
  • ✓Priority support (SLA 24 h)
  • ✓All Free tier features
Upgrade to Pro

Scale

$200 / mo

Long-running pipelines and high-throughput fleets.

  • ✓2 hr idle timeout
  • ✓24 hr max lifetime
  • ✓16 GB RAM • 8 vCPU • 64 GB disk per sandbox
  • ✓500 concurrent sandboxes
  • ✓SLA 4 h response
  • ✓All Pro tier features
Upgrade to Scale

Enterprise

Custom

Custom regions, dedicated infra, SOC 2 on request.

  • ✓Unlimited session length
  • ✓Custom RAM / vCPU / disk
  • ✓Dedicated hostd pool — no noisy neighbors
  • ✓Named engineer on Slack
  • ✓SOC 2 Type II available
  • ✓Data residency + private endpoints
Contact sales

How billing works

Monthly subscription

Pro and Scale include your base capacity. Overage beyond tier limits rolls to pay-as-you-go compute at published rates.

Pay-as-you-go compute

$0.05 / vCPU-hour · $0.016 / GiB RAM-hour · $0.09 / GB egress. Metered by the second; billed monthly.

Free credit

New accounts get $200 in credit — enough to prototype a full agent integration. Credit applies to both subscription and usage fees.

FAQ

+Can I upgrade or downgrade at any time?
Yes. Upgrades apply immediately; downgrades apply at the end of the current billing period. Limits are enforced server-side at create time — existing long-running sandboxes are never terminated by a tier change.
+What counts as a 'concurrent sandbox'?
Any sandbox whose destroy has not yet been called — includes idle sandboxes still in the warm window. Reaped sandboxes (idle timeout / max lifetime) drop out of the count immediately.
+Do you support SOC 2 / HIPAA / data residency?
On Enterprise only. Email sales@podflare.ai with your requirements.
+How fast is 'blazing fast'?
Server-side warm-pool restore is ~12 ms — the Podflare microVM is already booted when your request arrives. End-to-end round trip from a US laptop to us-west is p50 258 ms including TLS and the create + exec + destroy HTTPS calls. Numbers, and the bench that produced them, are on the Performance page.

Ready to ship?

Create an account, mint an API key, and run code in a sandbox in under a minute.

Sign up free →