AI Agent
Pricing Index
How much does an AI agent cost? At current list prices, a 3-minute AI voice call works out to about $0.26, roughly $0.09 a minute. The tables below hold the published price of every component behind that number, tokens, voice minutes, speech, and orchestration, across per-token, per-minute, and per-seat pricing models. Each figure links to the vendor page it was read from, on the date it was read.
LLM API layer
| Product | Input | Output |
|---|---|---|
Claude Fable 5.1 Anthropic Replaces Fable 5, which stays available as a legacy model at the same price. platform.claude.com, source for Claude Fable 5.1 pricing, opens in a new tab | $10.00 | $50.00 |
Claude Opus 5.5 Anthropic Replaces Opus 5, which stays available as a legacy model at $5.00 / $25.00. platform.claude.com, source for Claude Opus 5.5 pricing, opens in a new tab | $4.00 | $20.00 |
Claude Sonnet 5.5 Anthropic Legacy Sonnet 5 is also $2.00 / $10.00: Anthropic made its introductory rate permanent and cancelled the rise to $3.00 / $15.00 it had scheduled for 2026-09-01. platform.claude.com, source for Claude Sonnet 5.5 pricing, opens in a new tab | $2.00 | $10.00 |
Claude Haiku 5.5 Anthropic Prompts up to 100,000 tokens; $0.50 / $2.50 above that. Replaces Haiku 4.5, which stays available as a legacy model at $1.00 / $5.00. platform.claude.com, source for Claude Haiku 5.5 pricing, opens in a new tab | $0.10 | $0.50 |
gpt-6-astra OpenAI Standard tier, short context, as are the other GPT-6 rows. developers.openai.com, source for gpt-6-astra pricing, opens in a new tab | $10.00 | $50.00 |
gpt-6.1-sol OpenAI developers.openai.com, source for gpt-6.1-sol pricing, opens in a new tab | $2.00 | $10.00 |
gpt-6-luna OpenAI developers.openai.com, source for gpt-6-luna pricing, opens in a new tab | $0.10 | $0.50 |
GPT-5.6-sol OpenAI Standard tier, short context; long-context prompts are billed higher. Promotional pricing the vendor says runs at least through November 21, 2026. Was $5.00 / $30.00 when read on 2026-08-03. developers.openai.com, source for GPT-5.6-sol pricing, opens in a new tab | $4.00 | $20.00 |
GPT-5.6-terra OpenAI developers.openai.com, source for GPT-5.6-terra pricing, opens in a new tab | $2.00 | $12.00 |
GPT-5.6-luna OpenAI developers.openai.com, source for GPT-5.6-luna pricing, opens in a new tab | $0.20 | $1.20 |
Gemini 3.1 Pro Preview Google Prompts at or under 200k tokens; $4.00 / $18.00 above that. ai.google.dev, source for Gemini 3.1 Pro Preview pricing, opens in a new tab | $2.00 | $12.00 |
Gemini 3.8 Flash Google Price runs through 2026-12-31; $1.50 / $7.50 from 2027-01-01. ai.google.dev, source for Gemini 3.8 Flash pricing, opens in a new tab | $0.75 | $3.75 |
Gemini 3.5 Flash-Lite Google ai.google.dev, source for Gemini 3.5 Flash-Lite pricing, opens in a new tab | $0.30 | $2.50 |
DeepSeek-V4.1-Flash DeepSeek API name deepseek-flash. Off-peak cache-miss input / off-peak output; peak doubles both ($0.30 / $1.20). Peak is 01:00-04:00 and 06:00-10:00 UTC, Monday to Friday. Cache-hit input is $0.003 off-peak. Replaces DeepSeek-V4-flash, which was $0.22 / $0.66 off-peak on 2026-09-02; the old name is still accepted and served by V4.1-Flash. api-docs.deepseek.com, source for DeepSeek-V4.1-Flash pricing, opens in a new tab | $0.15 | $0.60 |
DeepSeek-V4-pro DeepSeek Off-peak cache-miss input / off-peak output. Peak doubles both ($1.32 / $3.96). Cache-hit input is $0.022 off-peak. Was a flat $0.435 / $0.87 on 2026-08-03; the peak-hour pricing this row previously flagged as announced is now live. api-docs.deepseek.com, source for DeepSeek-V4-pro pricing, opens in a new tab | $0.66 | $1.98 |
Voice agent platform
| Product | Platform fee |
|---|---|
Vapi Vapi Hosting only. STT, LLM and TTS are passed through at cost. vapi.ai, source for Vapi pricing, opens in a new tab | $0.05 |
Retell AI Retell AI Voice infrastructure only. Vendor quotes $0.07–$0.31/min all-in depending on configuration. retellai.com, source for Retell AI pricing, opens in a new tab | $0.055 |
Speech components
| Product | Price |
|---|---|
Flash / Turbo TTS ElevenLabs ~75ms latency tier. elevenlabs.io, source for Flash / Turbo TTS pricing, opens in a new tab | $0.04 / 1,000 characters |
Multilingual v2/v3 TTS ElevenLabs v3 Conversational is listed separately at $0.04 / 1,000 characters, ~280ms latency. elevenlabs.io, source for Multilingual v2/v3 TTS pricing, opens in a new tab | $0.08 / 1,000 characters |
Scribe v2 STT ElevenLabs elevenlabs.io, source for Scribe v2 STT pricing, opens in a new tab | $0.22 / hour |
Scribe v2 Realtime STT ElevenLabs ~150ms latency, word-level timestamps. elevenlabs.io, source for Scribe v2 Realtime STT pricing, opens in a new tab | $0.39 / hour |
Workflow automation
| Product | From |
|---|---|
Make: Core Make 10,000 credits/mo, billed annually; $10.59/mo on monthly billing (both toggle states read in-browser 2026-10-11). Free tier gives 1,000 credits/mo. make.com, source for Make: Core pricing, opens in a new tab | $9/mo |
Zapier: Professional Zapier Annual billing, "starting from" price. Monthly billing starts at $29.99. zapier.com, source for Zapier: Professional pricing, opens in a new tab | $19.99/mo |
n8n: Starter n8n Billed annually, 2,500 workflow executions. Priced in EUR by the vendor. n8n.io, source for n8n: Starter pricing, opens in a new tab | €20/mo |
n8n: Pro n8n Billed annually, 10,000 workflow executions. n8n.io, source for n8n: Pro pricing, opens in a new tab | €50/mo |
What one call actually costs
One 3-minute inbound call, answered by a voice agent. This is a model, not a measurement, every input is a published price from the tables above, and every assumption is listed so you can substitute your own.
| Component | Calculation | Cost |
|---|---|---|
| Vapi platform hosting | 3 min x $0.05 | $0.150 |
| STT: ElevenLabs Scribe v2 | 0.05 h x $0.22/h | $0.011 |
| TTS: ElevenLabs Flash | 1,350 chars x $0.04/1,000 | $0.054 |
| LLM: Claude Haiku 5.5 | 4,000 x $0.10/M + 800 x $0.50/M | $0.001 |
| Telephony | 3 min x $0.015 (Retell's published US rate) | $0.045 |
| Total | ≈ $0.09/minute | $0.26 |
Assumptions
- Call lasts 3 minutes end to end.
- The agent speaks for roughly half of it (~1.5 min); the caller speaks the rest.
- Speech runs ~150 words/min at ~6 characters per word, so the agent generates ~1,350 characters.
- The LLM sees ~4,000 input tokens (system prompt, tools, running transcript) and returns ~800 output tokens across the call.
- Cheapest published component in each tier is used, with no prompt caching and no volume discount.
Sanity check
This lands inside Retell AI's own published $0.07–$0.31/min all-in range, which is a useful sanity check on the model rather than an independent confirmation.
Methodology
- What this is
- A dated snapshot of publicly published list prices for the components used to build an AI voice or chat agent, with a link to the vendor page behind every number.
- What this is not
- Not a survey, not a benchmark, and not negotiated or measured pricing. We did not test these products to produce these figures, and we have no sample of real customer invoices. Enterprise and volume rates are routinely lower than list.
- How the figures were collected
- Each vendor's public pricing page was opened directly and the figure copied as published. Where a vendor publishes a range or a tiered rate, the row records the tier and the caveat rather than flattening it to one number.
- Limitations
- List prices change without notice; treat any figure as accurate only as of the verification date.
- Currencies are shown as the vendor publishes them and are not converted.
- Per-minute voice totals depend heavily on how much the agent speaks, which varies by script.
- Cache hits, batch processing and committed-use discounts can cut LLM cost substantially and are excluded from headline rows.
- Citing this page
- These figures are free to cite or reproduce with attribution and a link to this page.
Model your own volume
Swap in your own call length, speaking ratio, and monthly volume to see how these unit costs add up.
Project your costsChoosing between platforms rather than pricing one? See the AI agent pricing comparison : 20+ platforms compared head-to-head with ROI math.
Use this data
Published under CC BY 4.0. Reproduce any table, in whole or in part, commercially or otherwise attribution is the only condition. Every figure is a list price read from the vendor's own pricing page on October 11, 2026, and each row links to the page it came from, so anything here can be checked rather than trusted.
Cite it as:
PxlPeak. “AI Agent Pricing Index.” Prices verified October 11, 2026. https://pxlpeak.com/research/ai-agent-pricing-index
Prices move. If you are citing a figure, cite the verification date with it this page is re-checked against the vendor pages rather than left to age.
Ready?
Put AI to Work for You.
Call now and talk to Aria, our AI strategist, or book a free 30-minute assessment.
Aria picks up instantly · 24/7 · Free assessment · 30-day guarantee