AI Voice Agent pricing compared
What 4 ai voice agent tools actually cost — free tiers, the cheapest paid plan, and where each free plan runs out. Each figure is taken from the vendor’s own pricing page and dated, so you can see when we last checked.
| Tool | Free tier | Cheapest paid plan | Verified |
|---|---|---|---|
| Vapi | Yes | $0.05//minute · Pay as you go | 15 Jul 2026 |
| Retell AI | Yes | $0.07//minute · Pay as you go | 15 Jul 2026 |
| Bland AI | Yes | $299/mo platform fee · Paid tier 2 (name not shown on page) | 15 Jul 2026 |
| Synthflow | No | $30,000//year · Enterprise | 15 Jul 2026 |
Prices are as published by each vendor and may be region-specific — many vendors price by country. Confirm on the vendor’s page before buying.
Sorted cheapest paid plan first. A “free tier” means there is a genuinely free plan — tap through to any tool to see the full tier breakdown and exactly where the free plan stops. We list no fabricated prices: where a vendor only quotes “contact sales” we say so.
What drives ai voice agent pricing
Voice agents bill on talk time first and everything else second. The base meter is per minute of connected call, roughly $0.05 on Vapi, $0.07-$0.31 on Retell, and $0.11-$0.14 on Bland depending on tier, and that per-minute rate falls as you pay a higher monthly platform fee, since Bland's $0, $299, and $499 steps buy cheaper minutes and more concurrency. On top of the minute sit several pass-through and capacity costs: the LLM and voice model are often billed at cost or as a per-second voice charge, telephony is separate, and concurrency, the number of simultaneous calls, is capacity you rent, with Vapi including ten lines then charging $10 per extra line a month. Compliance add-ons like HIPAA and zero-data-retention carry their own monthly fees. Synthflow skips the menu with enterprise contracts from $30,000 a year. The bill is minutes times rate, plus concurrency, plus pass-through, plus platform fee.
How to choose on price
Model your call minutes and peak concurrency, because those two numbers dominate everything else. Estimate monthly connected minutes, then compare blended per-minute cost rather than headline rates, since Retell's $0.07-$0.31 range and Vapi's $0.05 land very differently once you add the voice and LLM you actually use. If volume is high, a platform fee that lowers the per-minute rate pays for itself: Bland's $299 or $499 tiers cut the talk rate and unlock more concurrent lines, which matters the moment inbound calls overlap. Bringing your own model API key, supported on Vapi, removes the model markup for heavy users. Watch capacity: under-provisioning concurrency drops calls, while over-provisioning rents idle lines at $10 each. Reserve Synthflow's $30,000-a-year contract for genuine scale. The false economy is chasing the lowest per-minute number while ignoring the concurrency, telephony, and model pass-through that make up most of a real bill.
AI Voice Agent pricing FAQ
What actually makes up a voice-agent bill?
Mostly per-minute talk time, but not only that. On top of the minute, roughly $0.05-$0.31 across these tools, sit pass-through model and voice costs, separate telephony, rented concurrency lines, and any platform fee. Compliance add-ons like Vapi's HIPAA at $2,000/mo stack on top. Budget the whole menu, not just the headline rate.
Why does a higher monthly platform fee sometimes cost less overall?
Because it buys cheaper minutes and more concurrency. Bland's $299 and $499 tiers lower the per-minute talk and transfer rates and raise simultaneous-call limits versus the $0 plan. Past a certain call volume, the saved per-minute cost outweighs the fixed fee, so heavy users pay less by paying more upfront.
What is concurrency and why do I pay for it?
Concurrency is the number of calls your agent can handle at once. It is rented capacity: Vapi includes ten lines and charges $10 per additional line a month, and Bland's tiers raise the limit from 10 to 100. Too few lines drops overlapping calls; too many wastes money on idle capacity.
Can I reduce cost by bringing my own model API key?
Yes, on platforms that pass model costs through. Vapi charges $0 for the model if you supply your own API key, removing that markup, while otherwise LLM and voice costs are billed at cost or per second. For high call volume, owning the model relationship can meaningfully cut the per-minute total.
See the full breakdown
Open any tool for its complete pricing tiers, free-tier limits, screenshot and honest review.