How can I compare per-call, per-result and per-second tool pricing?
Compare AI agent tool prices by the units a task consumes.
By Agentik · Updated
Normalize tool prices around the deliverable you need. One call can return several records, one image can have size adjustments, and a browser task can report actual consumed cost. Agentik publishes the selected operation’s pricing contract so your agent can distinguish the reservation maximum from the final charged units.
On this page
Read the unit before multiplying the rate
Per-call pricing charges an operation according to its contract. Per-unit pricing can measure returned records, seconds, characters or provider credits. Reported-cost tools use declared provider cost fields and their bounded reservation rules where available.
Inspect defaults, rounding, multipliers and maximums with the actual input. A requested result limit does not always equal billable units. A provider credit is not necessarily one record, and changing output size or audio settings can alter the quote.
Compare equivalent outputs and acceptance criteria
For lead enrichment, compare accepted records with the required fields and verification states. For media, compare the same dimensions, duration and audio requirements. A rate comparison is misleading when one output includes useful fields the other lacks.
Record total settled usage for the authorized task, including partial work and necessary retries. Keep model bills, funding fees, taxes and existing subscriptions separate. No savings percentage should be claimed without the underlying records and comparable scope.
Keep unknown or unbounded cases explicit
When the catalog cannot provide a supported maximum for the selected input, ask for a quote or an acceptable alternative rather than inventing a number. A zero displayed estimate is not an appropriate substitute for an unknown price.
A key spending cap also depends on the tool’s supported reservation contract. Do not bypass a configured cap to make an actual-cost operation run. Review the pricing and policy together before proceeding.
What you pay for
Agentik uses a prepaid workspace balance. Supported provider tools are billed for measured usage at the provider price, with no markup on tool calls. Funding the balance can carry a separate platform fee.
The published standard top-up fee is 5.5%, with a minimum of $0.80. The fee is added to the requested credit. The minimum top-up is $5.00. Check the Pricing page for any first-top-up waiver and the payment screen for the applicable fee and tax.
A call can reserve its maximum before dispatch, settle what it consumed and release the unused part. A partial result or an interrupted connection can still have consumed provider resources. Follow the existing run and its receipt rather than starting another paid call.
Your assistant or model provider bills its own usage separately. Existing credits at another provider are not transferred into Agentik. Connecting your own private accounts can require their own permissions or plans.
Normalize three different task units
These are calculation rules, not provider prices. Read the live contract for each selected implementation.
On a small screen, scroll to see every column.
| Operation | Inputs to normalize | Comparison unit |
|---|---|---|
| Email verification | Requested addresses and actual charged status units. | Cost per accepted checked record. |
| Voiceover | Final text characters and supported settings. | Cost for the approved script. |
| Video | Seconds, audio mode, size and attempts. | Cost for the accepted clip specification. |
A prompt to try
Connect your agent, then use this prompt with your connected assistant. Keep credentials in the authorized connection.
Give this to your agent
Compare the pricing of [tools] for [same deliverable]. Use Agentik inspect_tool to read rates, units, defaults, rounding, multipliers and maximums for the proposed inputs. Show each reservation calculation and the output differences. Keep unknown maxima explicit, separate funding and model costs, and define how settled cost per accepted deliverable will be measured after an approved trial. Do not invent savings or substitute zero for an unknown price.
Your agent should inspect the tools and ask for missing inputs before proposing any paid work.
A related workflow
Seed terms in, a ranked keyword list out, with what page one covers and how people ask in forums.
3 seed keywords for the United States in English, 100 ideas scored, page one read for the top 5 keywords, 5 forum searches.
This catalog-based reservation maximum belongs to the playbook’s inputs, separately from this guide’s worked example. It is not an executed charge. Optional calls, funding fees, taxes and model usage are excluded.
- Google keyword ideasDataForSEO
- Google bulk keyword difficultyDataForSEO
- Google search intentDataForSEO
Questions
Is the reservation the final tool charge?
No. It bounds or holds credit according to the contract. Settlement follows the actual pricing and consumed usage.
Can I compare tools using only price per call?
Only when the operations and outputs are equivalent. Normalize result limits, fields, duration, quality settings and expected accepted output.
Sources and next steps
Agentik publishes this guide. References below describe the product, its billing and the alternatives discussed. This page was updated on ; provider terms and tool availability can change.
Continue with Agentik
Related guides
- One balance for your agent's tool callsFollow the cost from the proposed operation to its settled receipt.
- Reduce the cost of your agent's tool workEstimate equivalent usage, limit duplicate calls and compare the complete cost.
- Compare providers for the same agent taskUse matching inputs and inspect the output you actually need.