Question 1

Is GPU CLI really free?

Accepted Answer

Yes. GPU CLI is free forever, no account required. Bring your own GPU or a provider API key. You pay the provider directly at their rates, zero markup. Every command, every provider integration, every workflow feature ships in the free CLI, with no tier-gated features, no usage caps from us, and no upsell paywalls inside the tool. Separate paid products like managed inference are on the pricing roadmap.

Question 2

Will auto-stop kill my training?

Accepted Answer

No. Auto-stop only triggers after your command completes. Active GPU work is never interrupted. Your training runs to completion, then auto-stop kicks in five minutes after the last activity.

Question 3

What happens when I close my laptop?

Accepted Answer

Your job keeps running. Reconnect anytime to see progress. Output files sync to your machine automatically as they are created.

Question 4

How is this different from runpodctl?

Accepted Answer

runpodctl is great for RunPod infrastructure management. GPU CLI adds a development workflow layer across providers: instant relay connections, automatic file syncing, and auto-stop to keep costs down.

Question 5

Can you see my code?

Accepted Answer

No. Your code syncs directly to your GPU provider over SSH or secure tunnels. We only see operational metadata such as duration, GPU type, and cost. Never your actual code or outputs.

Question 6

How much does a typical run cost?

Accepted Answer

It depends on your provider and the GPU you choose. On common providers, a two-hour A100 run is often around $4. You pay the provider directly at their rate. GPU CLI itself is free.

Question 7

What about paid services?

Accepted Answer

Paid services for teams are on the roadmap: managed model endpoints, team workspaces, hard spend caps, and priority support. They will be additions for teams that need more, never paywalls around features you already use.

RTX 4500 Ada Generation

Recent prices

Live prices by provider

Run on this GPU