You are billed for everything the model reads, re-sent every turn. Where the cost really goes, why agents multiply it, and the model-selection decision that fixes most of it.
- The cost is not ten times a single call — it is more.
- API cost runs 30–50% of margin on agency-style AI work, and voice runs $0.12–0.30 per minute all-in once speech-to-text, text-to-speech and telephony are counted.
You are billed for everything the model reads and everything it writes — not just your question. A long conversation means re-sending the whole conversation every single turn.
That is why costs climb without your usage changing.
Where the money actually goes
Context is re-sent every turn
Turn twenty of a conversation sends turns one through nineteen again. A long thread costs quadratically, not linearly — and nothing in the interface tells you this is happening.
The fix is unglamorous: start a fresh conversation when the topic changes. It is also better for accuracy — see loops that degrade.
Attached material is charged in full, every turn
A document you attached once is in context for the rest of the session. Attaching five reference files to answer one question about one of them pays for all five, repeatedly.
Agents multiply everything
An agent reads to decide what to read next. Ten steps means ten context loads, each larger than the last as the trace accumulates. The cost is not ten times a single call — it is more.
Multi-agent multiplies again, per handoff. Which is the economic argument on top of the reliability one in multi-agent.
Retries are silent
A failed call that retries three times bills three times. Loops that reprompt on dissatisfaction bill every pass. The runs you abandoned still cost what they cost.
The cheapest fix is usually the model
Frontier models are priced for frontier tasks. Most of what people run through them is not.
- Classification, extraction, formatting, routing — a small model handles these at a fraction of the cost and often faster.
- Reasoning, synthesis, judgement, long-context work — worth the frontier price.
- The test: would a competent junior do this correctly with the instructions you gave? If yes, a small model probably will too.
Most cost problems are model-selection problems wearing a prompt-optimisation costume.
What actually reduces spend
- Shorter context, not shorter prompts. The prompt is rarely the expensive part.
- Fresh conversations. The single highest-leverage habit.
- Cap agent steps. An uncapped agent is an uncapped bill.
- Cache what does not change. Where a provider supports it, a stable system brief is far cheaper re-used than re-sent.
- Cap the retries. Two attempts, then stop and report — see the agent brief.
- Ask it to check rather than regenerate. Checking is cheaper and more reliable.
If you are billing a client for this
The advertised per-minute rate is a floor, not a price. Retell publishes $0.07/min and Vapi $0.05/min — but those are base platform rates. Once speech-to-text, an LLM, text-to-speech and telephony are stacked on, the real cost lands at $0.13–$0.33.Verified 22 Aug 2026 against Autocalls, Aircall, Retell, Trillet, Kommunicate and WhiteLabelAI pricing analyses, Mar–Aug 2026
Reported agency economics: cost to run an agent $0.10–$0.20/min, clients billed at a $297–997 monthly retainer with 50–70% margins, or $300–1,500/month plus a $500–3,000 one-off setup.Trillet, Jan 2026 · Kommunicate, Aug 2026 · Suyash Raj, May 2026
A retainer priced off the advertised rate is a retainer priced off a number that does not exist. The gap between $0.05 and $0.33 is roughly six-fold, and it is entirely made of things the pricing page does not list.
A retainer priced without the meter is a retainer that gets thinner the more the client uses it.