Question 1

What am I actually paying for?

Accepted Answer

Two things: infrastructure (CPU, RAM, storage, network) and LLM tokens. Infrastructure is billed per hour at AWS-equivalent rates — no markup. LLM tokens are billed at the AI provider's cost with no surcharge. Your bill itemizes each resource separately.

Question 2

Do I get free credit to start?

Accepted Answer

Yes. Every new account gets $30 in free credit, no card needed. The credit is valid for 14 days. After that there is no subscription: you deposit credits and pay only for what your agents actually use, topping up any time.

Question 3

What happens when my agent is paused?

Accepted Answer

When you pause an agent, CPU and RAM charges stop immediately. The only ongoing cost is storage — you're still holding the agent's data on disk. LLM tokens are only consumed when the agent is actively processing tasks, so those stop too.

Question 4

How are LLM token costs calculated?

Accepted Answer

LLM tokens are billed at the AI provider's published rates — GPT-5.3-Codex, Claude Opus 4.7, Gemini 3.5, GLM-5.1, Kimi K2.6, and more. There is no markup. If OpenAI charges $2.50 per 1M input tokens, that's exactly what you pay. Your dashboard shows token costs broken down by model.

Question 5

Can I bring my own API keys?

Accepted Answer

Yes. If you configure your own provider API keys, LLM token costs go to zero on your JackHamr bill — you pay the provider directly. You'll still pay for infrastructure (CPU and RAM while running, storage always) at the same AWS-equivalent rates.

Question 6

Can I change my agent size later?

Accepted Answer

Yes. You can scale your agent up or down at any time. The cost adjusts immediately based on the new resource allocation. No commitments, no reserved instances.

Pricing that doesn't punish you for scaling.

Cost Estimator

How billing works

Pay only while running

Per-resource transparency

No commitments

Frequently Asked Questions