AI agent cost calculator
30-day month.
Default 18,000 — source: reelier benchmark (avg 18,390 in + 136 out per agent run, N=10).
Input-token list price; sources below.
Level-0 replay makes no LLM calls — by construction.
* Replay re-executes your tool calls; API/egress costs of the tools themselves still apply. Judgment steps that genuinely need a model are opt-in escalation, not replay.
How the math works
No model behind this page — it’s three multiplications:
- runs/month— your interval over a 30-day month: every 5 minutes = 8,640 runs, every 15 minutes = 2,880, hourly = 720, daily = 30.
- tokens/month = runs/month × tokens/run
- $/month = tokens/month ÷ 1,000,000 × $/MTok
Two deliberate simplifications, both of which understatethe bill: we price every token at the model’s inputrate (output tokens cost 5–6× more on every model listed, but tool-heavy agent runs are input-dominated — the benchmark run below was 18,390 in / 136 out), and we ignore retries, drift-fixing re-runs, and tool-result bloat. Your real bill is higher.
The 18,000-token default is the measured average from the reelier benchmark — a real npm-registry agent workflow run N=10 on Claude Haiku 4.5: avg 18,390 input + 136 output tokens, ~$0.019/run. Your workflows may be bigger or smaller; that’s what the field is for.
Why the replay column says $0.00
A Reelier skill is the recorded run compiled into a file. Level-0 replay re-executes the tool calls and checks the assertions — it never calls a model, so LLM spend is zero by construction, not by discount. This is verified from the run record on every replay (llmInputTokens === 0), not assumed. The honest asterisk: replay re-executes your tool calls, so API/egress costs of the tools themselves still apply, and judgment steps that genuinely need a model are opt-in escalation, not replay.
npm i -g reelier && reelier init— record the run once, replay it on a schedule, keep the receipt.
Price sources
Input-token list prices, retrieved 2026-07-21. Claude Haiku 4.5 $1.00/MTok in ($5.00 out) and Claude Sonnet 4.6 $3.00/MTok in ($15.00 out): platform.claude.com/docs/en/about-claude/pricing. GPT-5.6 Sol $5.00/MTok in ($30.00 out): platform.openai.com/docs/pricing. Prices change — if these look stale, use the custom field and the current pages.