Pricing

Pay for what your agent does. Nothing while it rests.

No subscription, no seats — prepaid credits spent on two simple meters, itemized down to the turn. Signing up grants a $3 credit, no card required, and parked agents are free forever.

Harness

The agent software your agent runs — pick one at launch, swap anytime without losing its history.

Claude CodeCodexHermesOpenClawFree

Model usage

Billed per turn for the tokens your agent actually consumes, at each vendor's publicly listed price. Swap models anytime.

ModelInput / 1M tokensOutput / 1M tokens
Claude Fable 5$10.00$50.00
Claude Opus 4.8$5.00$25.00
Claude Sonnet 4.6$3.00$15.00
GPT-5.6 Sol$5.00$30.00
GPT-5.6 Terra$2.50$15.00
GPT-5.6 Luna$1.00$6.00
GPT-5.5$5.00$30.00
DeepSeek V4 Pro$0.44$0.87
DeepSeek V4 Flash$0.14$0.28
Gemini 3.5 Flash$1.50$9.00
GLM-5.2$1.40$4.40

Compute

Billed per second while your agent's machine is awake. It suspends automatically when idle and wakes on the next message.

MachinePer hour awakePer 24h awakeAlways on, per month
1 GB$0.021$0.51$15.24
2 GB$0.038$0.91$27.16
4 GBdefault$0.071$1.70$51.01
Suspended or parked$0 — idle agents suspend automatically; parked agents are free forever

Billed per second, only while awake; slivers under a minute aren't booked. Most agents sleep most of the day — the always-on column is the ceiling, not the norm.

Estimate your monthly cost

Assumes a typical turn of ~8k input + 1k output tokens. Idle time suspends automatically and costs nothing — "hours awake" is time actually spent working.

Estimated monthly cost

$82.25

$78.00 model usage$4.25 compute

Your free $3 credit covers the first ~1 day.

Spend $0 finding out

Your $3 welcome credit covers the whole experiment: launch an agent on any harness, put it to work, and park it free if it's not for you. No card, no plan, no lock-in.