Question 1

What does active CPU hour mean?

Accepted Answer

You are billed based on how much CPU your code actually consumes, measured in core-hours. For example, using 100% of 2 cores for one hour costs $0.2, while using 10% of a single core for one hour costs $0.01. When a standard box is idle with no CPU usage, no charges apply. Keep-alive boxes use fixed monthly pricing by size instead.

Question 2

What happens when a box is idle?

Accepted Answer

Standard boxes automatically freeze after inactivity. Storage is preserved, but running processes are stopped until the next request. When a new request comes in, the box wakes up automatically, similar to how AWS Lambda cold-starts on demand. No manual intervention is needed. Keep-alive boxes stay on continuously and use fixed monthly pricing by size instead of auto-pause behavior.

Question 3

How much do Keep Alive boxes cost?

Accepted Answer

Keep Alive boxes are billed with a fixed monthly price by size. This is the only charge for that Keep Alive box, so you are not billed separately for CPU or storage on it. Small costs $8/month, Medium costs $16/month, and Large costs $32/month.

Question 4

What is included in storage billing?

Accepted Answer

Storage is billed at $0.1 per GB per month and includes all data on disk, including snapshots. Storage limits are 5 GB for Small, 10 GB for Medium, and 20 GB for Large.

Question 5

What are the free tier limits?

Accepted Answer

The free tier includes up to 10 concurrent boxes, 5 CPU hours per month, and $1 of LLM token usage per month. After the token limit, the API returns 400 Bad Request.

Question 6

How does LLM token usage work?

Accepted Answer

Each box includes Upstash-provided LLM token usage for the built-in AI agent. On the Free plan, this is capped at $1 per month. On Pay as you go, the cap is $100 per month. After the limit is reached, you can either contact us to increase it or bring your own key (BYOK) for any LLM provider, which is available on all plans.

Question 7

Can I run more than 100 concurrent boxes?

Accepted Answer

Yes. The default quota for Pay as you go is 100 concurrent boxes. This is a soft limit that can be increased on request, similar to AWS service quotas. Just contact us and we'll raise your limit.

Question 8

Where are boxes hosted?

Accepted Answer

Boxes currently run on AWS us-east-1. We are planning to add more regions soon.

Question 9

Can I use my own Docker container?

Accepted Answer

Yes, custom Docker container support is planned and coming soon.

Capacity
	Free	Pay as you go	Enterprise
	Free -	$0.1/ active CPU hour	Enterprise Custom
	10	1000 (default)	Custom
CPU Cores / Box	2	2	Custom
Memory / Box	4GB	4GB	Custom
Max Storage / Box	5GB	5GB	Custom
CPU Hours / Month	5	Unlimited	Unlimited
	30 minutes	1 hour	Custom
Keep Alive
Runtimes	Node.js, Python, Go, Ruby, Rust	Node.js, Python, Go, Ruby, Rust	Node.js, Python, Go, Ruby, Rust
		Coming Soon	Coming Soon

Price
Monthly Price	Free	None	Custom
	Free	$0.1 / active CPU hour	Custom
Memory Price	Free	Free	Custom
Storage Price	Free	$0.1 / GB per month	Custom
	Start Now	Start Now	Contact Us

Pricing

Free

$0

Pay as you go

$0.10

Enterprise

Custom

Compare Plans

Free

-

$0.1/ active CPU hour

Enterprise

Custom

FAQ

Free

$0

Pay as you go

$0.10

Enterprise

Custom

Free

-

$0.1/ active CPU hour

Enterprise

Custom

FAQ

What does "active CPU hour" mean?

What happens when a box is idle?

How much do Keep Alive boxes cost?

What is included in storage billing?

What are the free tier limits?

How does LLM token usage work?

Can I run more than 100 concurrent boxes?

Where are boxes hosted?

Can I use my own Docker container?