Coding · pricing
Choose AgentSky if you want to run Claude Code, Codex or other coding agents in a browser or through one API and pay only for compute seconds and model tokens, or $0 model cost on a connected Claude or ChatGPT plan.
AgentSky has no seats and no minimums; you prepay credits and are billed by usage. Compute is $0.461 per hour on the default 2 vCPU, 8 GB machine, metered per second while an agent runs, and a parked agent bills nothing. Model tokens are billed per 1M tokens at each model's rate, giving two examples in the source: Claude Sonnet 5 at $2 input and $10 output, and GPT-6 Luna at $0.10 input and $0.50 output, through one key to 16+ models.
If you connect a Claude Pro or Max plan, Claude Code agents bill model usage at $0 to your own subscription; a ChatGPT Plus or Pro plan does the same for Codex agents. Compute is still metered either way. No free tier is confirmed. Prices checked on 2026-10-01.
| Plan | Price | Billing | Who it is for | Key limits | API |
|---|---|---|---|---|---|
| Pay as you go - compute web |
$0.46 per hour (2 vCPU, 8 GB) | Hour | Occasional hosted agent runs | Prepaid credits; Metered per second while agent runs | Included |
| Pay as you go - model tokens web |
Usage-based | — | No existing AI subscription | Example: Claude Sonnet 5 $2 input / $10 output per 1M; Example: GPT-6 Luna $0.10 / $0.50 per 1M | Included |
| Connected Claude or ChatGPT plan web |
$0 | — | Existing Claude or ChatGPT subscribers | Claude Pro/Max covers Claude Code agents; ChatGPT Plus/Pro covers Codex agents; Compute still metered | Included |
| If you | Pick | Why |
|---|---|---|
| you run hosted agents only occasionally | Pay as you go - compute, $0.461/hour | No seats or minimums, billed per second only while the agent runs. |
| you have no AI subscription and want many models behind one key | Pay as you go - model tokens (per 1M tokens, per model) | Each of 16+ models bills at its own published rate, on top of compute. |
| you already pay for Claude Pro or Max, or ChatGPT Plus or Pro | Connected Claude or ChatGPT plan ($0 model usage) | Tokens bill to your own subscription; you only pay for compute and extras. |
No free tier confirmed; model usage is $0 on a connected Claude or ChatGPT plan, but credits are still needed for compute.
Connector calls and paid capabilities charged per use, e.g. $4 per connector (up to 4 apps, up to $16).
Prices checked Oct 1, 2026 on the official US pricing page. Official pricing ↗ Prices and limits change often.
No free tier confirmed; model usage is $0 on a connected Claude or ChatGPT plan, but credits are still needed for compute.
$0.46 per hour (2 vCPU, 8 GB) source (opens in a new tab)
Hour
For: Occasional hosted agent runs
Limits
Includes
API included
Usage-based source (opens in a new tab)
For: No existing AI subscription
Limits
Includes
API included
For: Existing Claude or ChatGPT subscribers
Limits
Includes
API included
Connector calls and paid capabilities charged per use, e.g. $4 per connector (up to 4 apps, up to $16).
Prepaid credits on two meters: compute per second (about $0.461/hour for 2 vCPU, 8 GB) and model tokens at each model's published rate. source (opens in a new tab)
Yes. Connect Claude Pro/Max for Claude Code agents or ChatGPT Plus/Pro for Codex agents and AgentSky bills $0 for model usage; compute is still metered. source (opens in a new tab)
No. A parked agent bills neither compute nor model tokens. source (opens in a new tab)
None is confirmed. Model usage is $0 on a connected Claude or ChatGPT plan, but credits are still needed for compute.
$0.461 per hour on the default 2 vCPU, 8 GB machine, metered per second while the agent runs. A parked agent bills nothing.
It depends on the model: for example Claude Sonnet 5 is $2 input and $10 output per 1M tokens, and GPT-6 Luna $0.10 and $0.50.
Yes. Claude Pro or Max covers Claude Code agents and ChatGPT Plus or Pro covers Codex agents, though compute is still billed.
compare>