01
Metered every 2 minutes
Usage is calculated in 2-minute micro-buckets across compute, memory, storage, bandwidth, and GPU time — no rounding up to full hours.
Pricing
Pay for active runtime.Not for idle hours.
Usage is calculated in 2-minute micro-buckets across compute, memory, and storage. Sleeping containers pause compute billing completely, and a $0 balance stops paid workloads rather than sending unexpected bills.
Project workflow01 / 03
01
Source360 free shared hours monthly
02
ControlNo compute costs while asleep
03
ReleaseNo surprise overages
Hubfly space projectNetwork · volumes · domains · team access
One project model
Rate card
Three simple rates. That's our entire pricing model.
Every workload on Hubfly space is priced using three hourly rates. No hidden user seat costs, no per-project fees, and no charge for attaching custom domains.
CPU
$0.004
per vCPU / hour
You're only billed while your container is awake.
RAM
$0.002
per GB / hour
Metered against exact memory allocated.
Storage
$0.00001536
per GB / hour
Works out to roughly $0.011 per GB each month.
Shared tier
360h
free runtime monthly + 1GB storage
Resets every month automatically, no credit card required.
Estimate
Estimate your container costs in seconds.
1
1 GB
10 GB
Estimated monthly cost
$6.05
$0.0083 per hour · 730 hrs
- Compute
- $2.92
- Memory
- $1.46
- Storage
- $0.11
This is a quick estimate for planning. Real billing is metered in exact 2-minute buckets based on active usage, meaning sleeping containers cost even less.
Worked examples
Real-world project cost estimates.
All figures below demonstrate our exact hourly rates multiplied over a standard 730-hour month.
Side project
0.25 vCPU · 512 MB · 1 GB disk
Shared, auto-sleep on
$0.00
per month
Fits completely inside your 360 free hours and 1 GB storage allowance.
Small production API
0.5 vCPU · 1 GB · 10 GB disk
Dedicated 24×7
≈ $3.03
per month
730 hrs × (0.5 × $0.004 + 1 × $0.002) + 10 GB disk storage.
Postgres + app stack
2 vCPU · 4 GB · 50 GB balanced
Dedicated 24×7
≈ $12.52
per month
Covers two full application containers plus a balanced managed volume.
Container tiers
Choose how your workloads handle sleep.
Container tiers control whether a workload scales to zero. Asleep containers release CPU and RAM, pausing compute billing until the next request wakes them up.
| Tier | Billing | Scales to zero | Best for |
|---|---|---|---|
| Shared Compute | Free tier eligible | Yes | Perfect for side projects, staging environments, and light APIs. |
| Dedicated Sleepable | Per vCPU + GB, while awake | Yes | Internal tools that only need to run during working hours. |
| Dedicated 24×7 | Per vCPU + GB, continuous | No | Production backends, databases, and WebSocket services that must stay up. |
| Auto-Scaling | Per vCPU + GB, scales with load | No | Apps with unpredictable traffic spikes needing live CPU, RAM, and disk scaling. |
| Docker-OS | Per vCPU + GB, continuous | No | Full virtual Linux environments running systemd with full system access. |
Shared Compute
- Billing
- Free tier eligible
- Scales to zero
- Yes
- Best for
- Perfect for side projects, staging environments, and light APIs.
Dedicated Sleepable
- Billing
- Per vCPU + GB, while awake
- Scales to zero
- Yes
- Best for
- Internal tools that only need to run during working hours.
Dedicated 24×7
- Billing
- Per vCPU + GB, continuous
- Scales to zero
- No
- Best for
- Production backends, databases, and WebSocket services that must stay up.
Auto-Scaling
- Billing
- Per vCPU + GB, scales with load
- Scales to zero
- No
- Best for
- Apps with unpredictable traffic spikes needing live CPU, RAM, and disk scaling.
Docker-OS
- Billing
- Per vCPU + GB, continuous
- Scales to zero
- No
- Best for
- Full virtual Linux environments running systemd with full system access.
Storage tiers
Storage options tailored to your performance needs.
Persistent volumes are billed using base storage rates multiplied by your selected tier profile. Storage capacity can grow live on active containers without downtime.
| Tier | Multiplier | Per GB / hour | Per GB / month | Typical use |
|---|---|---|---|---|
| Standard | 1.0× | $0.00001536 | $0.011 | Ideal for application logs, static files, backups, and low-traffic sites. |
| Balanced | 1.5× | $0.00002304 | $0.017 | Great for general-purpose databases and read-heavy workloads. |
| High Performance | 2.0× | $0.00003072 | $0.022 | Built for transactional Postgres and MySQL under heavy write operations. |
Standard
- Multiplier
- 1.0×
- Per GB / hour
- $0.00001536
- Per GB / month
- $0.011
- Typical use
- Ideal for application logs, static files, backups, and low-traffic sites.
Balanced
- Multiplier
- 1.5×
- Per GB / hour
- $0.00002304
- Per GB / month
- $0.017
- Typical use
- Great for general-purpose databases and read-heavy workloads.
High Performance
- Multiplier
- 2.0×
- Per GB / hour
- $0.00003072
- Per GB / month
- $0.022
- Typical use
- Built for transactional Postgres and MySQL under heavy write operations.
Monthly figures assume a 730-hour month.
How billing works
Transparent, auditable usage metering.
Usage streams from regional schedulers into raw buckets, logs to an immutable double-entry ledger, and renders into real-time metrics on your dashboard.
02
Micro-USD precision
Calculations run in micro-USD ($1.00 = 1,000,000 units), eliminating rounding errors so tiny charges are tracked accurately.
03
Promo credits spend first
Promotional credits and educational grants are automatically used up before your actual card balance is charged.
04
No surprise bills
If your balance reaches $0.00, paid workloads pause automatically. You'll never get hit with unexpected overage charges at month-end.
05
Itemised invoices
Every billing statement gives you a clean breakdown per container, volume, custom domain, and GPU instance.
06
Flexible payments
Pay via Stripe or Dodo Payments with manual top-ups or automatic recharge whenever your balance drops below your chosen threshold.
Zero-balance protection
Reaching $0 balance never results in surprise debt.
Instead of quietly charging your card for overages, Hubfly space gracefully pauses paid workloads and gives you a 3-day grace period to export data.
- 01
Balance reaches $0.00
Paid containers stop automatically to prevent further compute charges from piling up.
Immediate - 02
Three days at zero
Free containers and persistent storage turn read-only. Your data stays safe so you can export it anytime.
After 72 hours - 03
Balance reaches −$5.00
Resources are queued for permanent deletion. Email warnings go out well before anything gets removed.
Final step
Budget controls for teams and classes
Organizations can assign monthly, semester, or custom budget caps to project groups. Once hit, projects pause automatically to prevent runaway costs for student groups or internal experiments.
Allocation types
Monthly · Semester · One-time
Enforcement
Hard cap, projects halt
Metering window
2-minute buckets
Precision
Micro-USD (1e-6)
Questions
Pricing questions answered.
What counts against the 360 free hours?
Only runtime spent on the shared tier while your container is active. Since shared apps auto-sleep when idle, an unused project uses practically none of your allowance.
Do I pay while a container is asleep?
Nope. Asleep containers release CPU and RAM, so compute billing stops completely. You only pay for whatever persistent disk storage you keep.
How is GPU time billed?
GPU workloads are funded through a dedicated GPU wallet rather than your main account balance. You pay the raw hourly rate plus a simple markup.
What happens when a budget cap is hit?
Workloads linked to that budget freeze immediately once the cap is reached. It's a great safeguard for teams and classrooms to prevent runaway spending.
Start free
Get 360 free hours every month.
Deploy on our shared compute tier, let containers sleep when inactive, and scale up to dedicated tiers whenever your traffic grows.