{"name":"AI Agent Pricing Index","url":"https://conversionsystem.com/ai-agent-pricing-index","verifiedOn":"2026-08-27","license":"CC BY 4.0. Attribution to Conversion System with a link is enough.","method":"Every figure was read off the vendor's own published pricing page on the date in `verified`, and links back to it. Nothing is estimated. A null amount means the vendor publishes no number, which the note explains.","counts":{"pricedRows":45,"scheduledChanges":4,"researchFindings":4},"rows":[{"layer":"model","vendor":"Anthropic","product":"Claude Haiku 4.5","unit":"per 1M input tokens","amountUsd":1,"note":"Output is $5 per 1M. The cheapest current Claude tier, and the one Anthropic uses in its own support-agent cost example.","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"Anthropic","product":"Claude Sonnet 5","unit":"per 1M input tokens","amountUsd":2,"note":"Output is $10 per 1M. Anthropic states the $2/$10 launch pricing is now standard and the scheduled rise to $3/$15 will not happen.","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"Anthropic","product":"Claude Opus 5","unit":"per 1M input tokens","amountUsd":5,"note":"Output is $25 per 1M.","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"Anthropic","product":"Claude Fable 5","unit":"per 1M input tokens","amountUsd":10,"note":"Output is $50 per 1M. Top tier.","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"OpenAI","product":"gpt-5-nano","unit":"per 1M input tokens","amountUsd":0.05,"note":"Output is $0.40 per 1M. The cheapest row on the page.","source":"https://developers.openai.com/api/docs/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"OpenAI","product":"gpt-5.6-luna","unit":"per 1M input tokens","amountUsd":0.2,"note":"Output is $1.20 per 1M. Cached input is $0.02.","source":"https://developers.openai.com/api/docs/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"OpenAI","product":"gpt-5.6-terra","unit":"per 1M input tokens","amountUsd":2,"note":"Output is $12 per 1M. Cached input is $0.20.","source":"https://developers.openai.com/api/docs/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"OpenAI","product":"gpt-5.6-sol","unit":"per 1M input tokens","amountUsd":4,"note":"Output is $20 per 1M. The page states this promotional rate runs at least through 2026-11-21, so treat it as temporary.","source":"https://developers.openai.com/api/docs/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"Google","product":"Gemini 2.5 Flash-Lite","unit":"per 1M input tokens","amountUsd":0.1,"note":"Output is $0.40 per 1M. Audio input is priced higher at $0.30.","source":"https://ai.google.dev/gemini-api/docs/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"Google","product":"Gemini 3.5 Flash-Lite","unit":"per 1M input tokens","amountUsd":0.3,"note":"Output is $2.50 per 1M.","source":"https://ai.google.dev/gemini-api/docs/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"Google","product":"Gemini 3.7 Flash","unit":"per 1M input tokens","amountUsd":0.75,"note":"Output is $3.75 per 1M. The page dates a rise to $1.50 / $7.50 on 2027-01-01.","source":"https://ai.google.dev/gemini-api/docs/pricing","verified":"2026-08-27"},{"layer":"model","vendor":"Google","product":"Gemini 2.5 Pro","unit":"per 1M input tokens","amountUsd":1.25,"note":"Output is $10 per 1M including thinking tokens, and both roughly double above a 200k-token prompt.","source":"https://ai.google.dev/gemini-api/docs/pricing","verified":"2026-08-27"},{"layer":"inference-modifier","vendor":"Anthropic","product":"Prompt cache read","unit":"multiple of base input price","amountUsd":0.1,"note":"A cache hit costs 10% of standard input. Writes cost 1.25x (5-minute) or 2x (1-hour), so a 5-minute cache pays for itself after one read.","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"},{"layer":"inference-modifier","vendor":"Anthropic","product":"Batch API","unit":"discount on input and output","amountUsd":0.5,"note":"50% off both directions, for work that does not need an answer in real time. Not available for interactive sessions.","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"},{"layer":"inference-modifier","vendor":"Anthropic","product":"Web search tool","unit":"per 1,000 searches","amountUsd":10,"note":"Billed on top of tokens. An agent that searches on every turn adds a line item most cost models miss.","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"},{"layer":"inference-modifier","vendor":"Anthropic","product":"US-only inference (data residency)","unit":"multiplier on all token pricing","amountUsd":1.1,"note":"Pinning inference to the US adds 10% across input, output, and cache.","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"},{"layer":"platform","vendor":"Zapier","product":"Professional","meteredBy":"per task","unit":"per month, from","amountUsd":19.99,"note":"Printed as a starting price; the real figure moves with your task tier. Team starts at $69 per month for 25 users. A free tier gives 100 tasks a month.","source":"https://zapier.com/pricing","verified":"2026-08-27"},{"layer":"platform","vendor":"Make","product":"Core","meteredBy":"per credit (one module action)","unit":"per month","amountUsd":12,"note":"At the default 10,000 credits a month. Pro is $21 and Teams $38 at the same volume. Free tier allows up to 1,000 credits a month with no time limit.","source":"https://www.make.com/en/pricing","verified":"2026-08-27"},{"layer":"platform","vendor":"n8n","product":"Starter","meteredBy":"per full workflow execution, not per step","unit":"per month","amountUsd":null,"note":"Priced in euros even from a US request (EUR 20 a month billed annually, 2,500 executions), so no verified USD figure exists to quote. Metering per whole execution rather than per step is the notable part: a 30-step workflow costs the same as a 3-step one.","source":"https://n8n.io/pricing/","verified":"2026-08-27"},{"layer":"platform","vendor":"Lindy","product":"Plus","meteredBy":"per seat plus a shared credit pool","unit":"per user, per month","amountUsd":29.99,"note":"Includes 3,000 credits per user. Pro is $99.99 and Max $199.99 per user. No free tier.","source":"https://www.lindy.ai/pricing","verified":"2026-08-27"},{"layer":"platform","vendor":"Gumloop","product":"Pro","meteredBy":"per credit, plus an 8% orchestration fee","unit":"per month, from","amountUsd":37,"note":"The 8% fee stacks on credit consumption, so the sticker is not the effective rate. Only two tiers exist and there is no free plan.","source":"https://www.gumloop.com/pricing","verified":"2026-08-27"},{"layer":"platform","vendor":"Salesforce","product":"Agentforce, per action","meteredBy":"per action, via Flex Credits","unit":"per action","amountUsd":0.1,"note":"Derived from figures printed on the same page: an action is 20 Flex Credits and credits are $500 per 100,000. Voice actions are 30 credits, so $0.15. A per-conversation option is $2.","source":"https://www.salesforce.com/agentforce/pricing/","verified":"2026-08-27"},{"layer":"platform","vendor":"Microsoft","product":"Copilot Studio","meteredBy":"per Copilot Credit","unit":"per credit, pay as you go","amountUsd":0.01,"note":"A prepaid pack is $200 a month for 25,000 credits. Microsoft's own Azure pricing page renders this rate as a dash; the figure is confirmed in the Copilot Credits Guide.","source":"https://www.microsoft.com/en-us/microsoft-365-copilot/pricing/copilot-studio","verified":"2026-08-27"},{"layer":"platform","vendor":"Google","product":"Gemini Enterprise Agent Platform, compute","meteredBy":"per vCPU-hour","unit":"per vCPU-hour","amountUsd":0.085,"note":"Pure consumption, no seats and no tiers, with the first 50 vCPU-hours a month free. This product was called Vertex AI Agent Builder; the name no longer appears on the pricing page, so older comparisons are stale.","source":"https://cloud.google.com/vertex-ai/pricing","verified":"2026-08-27"},{"layer":"platform","vendor":"Relevance AI","product":"Self-serve plans","meteredBy":"Actions and Vendor Credits","unit":"per month","amountUsd":null,"note":"No self-serve price is published at all; the pricing page is a single Enterprise card with a sales contact. Only top-up rates remain documented, at $80 per 1,000 Actions.","source":"https://relevanceai.com/pricing","verified":"2026-08-27"},{"layer":"platform","vendor":"CrewAI","product":"Paid tiers","meteredBy":"per workflow execution","unit":"per month","amountUsd":null,"note":"Nothing purchasable sits between the free tier (50 executions a month) and a custom enterprise quote. There is no published middle.","source":"https://www.crewai.com/pricing","verified":"2026-08-27"},{"layer":"voice-platform","vendor":"Vapi","product":"Vapi hosting (Build)","unit":"per call minute","amountUsd":0.05,"note":"Platform fee only. The page states hosting excludes model provider costs: speech-to-text, the model, and speech synthesis are billed at cost, and transport is charged by the provider.","source":"https://vapi.ai/pricing","verified":"2026-08-27"},{"layer":"voice-platform","vendor":"Retell AI","product":"Voice infrastructure","unit":"per call minute","amountUsd":0.055,"note":"Retell is unusual in publishing a per-minute rate for every component, which is why its stack is the one that resolves exactly.","source":"https://www.retellai.com/pricing","verified":"2026-08-27"},{"layer":"voice-platform","vendor":"Retell AI","product":"Speech synthesis","unit":"per call minute","amountUsd":0.015,"note":"Rises to $0.040 per minute if you pick ElevenLabs voices.","source":"https://www.retellai.com/pricing","verified":"2026-08-27"},{"layer":"voice-platform","vendor":"Bland AI","product":"Start plan","unit":"per talk minute","amountUsd":0.14,"note":"Model, speech-to-text and speech synthesis are included with no token charges. Telephony is billed separately. The only vendor here whose headline number is close to the real one.","source":"https://www.bland.ai/pricing","verified":"2026-08-27"},{"layer":"voice-platform","vendor":"ElevenLabs","product":"Agents, additional minute","unit":"per call minute","amountUsd":0.08,"note":"Speech synthesis, speech-to-text and retrieval are included. The page states the model and any telephony are billed separately on top, both at cost.","source":"https://elevenlabs.io/pricing/agents","verified":"2026-08-27"},{"layer":"voice-platform","vendor":"Deepgram","product":"Voice Agent API (Standard)","unit":"per call minute","amountUsd":0.056,"note":"Bundles speech-to-text, model and speech synthesis orchestration, billed on websocket connection time. The page dates a rise to $0.075 on 2026-09-12. Telephony is not included.","source":"https://deepgram.com/pricing","verified":"2026-08-27"},{"layer":"voice-component","vendor":"Deepgram","product":"Nova-3 streaming speech-to-text","unit":"per minute","amountUsd":0.0048,"note":"Pay-as-you-go rate. Listening is the cheapest part of a voice agent by a wide margin.","source":"https://deepgram.com/pricing","verified":"2026-08-27"},{"layer":"voice-component","vendor":"Deepgram","product":"Aura-2 speech synthesis","unit":"per 1,000 characters","amountUsd":0.03,"note":"Charged per character, not per minute. This is the line that stops any stack from producing a clean per-minute number, because it depends on how much the agent says.","source":"https://deepgram.com/pricing","verified":"2026-08-27"},{"layer":"voice-component","vendor":"Twilio","product":"US local inbound voice","unit":"per minute","amountUsd":0.0085,"note":"Outbound US is $0.0140. SIP trunking, which is the rate most voice-AI platforms actually hit, is $0.0040 in both directions.","source":"https://www.twilio.com/en-us/voice/pricing/us","verified":"2026-08-27"},{"layer":"voice-component","vendor":"Twilio","product":"US local phone number","unit":"per month","amountUsd":1.15,"note":"Toll-free is $2.15 per month. A fixed cost per number, regardless of call volume.","source":"https://www.twilio.com/en-us/voice/pricing/us","verified":"2026-08-27"},{"layer":"voice-component","vendor":"OpenAI","product":"gpt-realtime-2.1 audio","unit":"per 1M audio input tokens","amountUsd":32,"note":"Audio output is $64 per 1M. OpenAI publishes NO per-minute price for realtime voice; it is token-billed only. Any \"$X per minute for OpenAI Realtime\" figure is somebody else's conversion.","source":"https://developers.openai.com/api/docs/pricing","verified":"2026-08-27"},{"layer":"labor","vendor":"US Bureau of Labor Statistics","product":"Data Scientists (SOC 15-2051)","measure":"median annual wage","basis":"Government occupational survey, OEWS May 2025. Read from the official BLS API and corroborated against O*NET to the dollar.","unit":"per year","amountUsd":120230,"note":"The 75th percentile is $158,880. Wages only: apply the employer cost loading below for a real comparison.","source":"https://www.bls.gov/oes/current/oes152051.htm","verified":"2026-08-27"},{"layer":"labor","vendor":"US Bureau of Labor Statistics","product":"Software Developers (SOC 15-1252)","measure":"median annual wage","basis":"Government occupational survey, OEWS May 2025.","unit":"per year","amountUsd":135980,"note":"The baseline AI roles get compared against. There is NO BLS occupation code for \"AI Engineer\", so any figure attributed to BLS under that title is invented.","source":"https://www.bls.gov/oes/current/oes151252.htm","verified":"2026-08-27"},{"layer":"labor","vendor":"US Bureau of Labor Statistics","product":"Employer cost loading","measure":"multiplier on wages","basis":"Employer Costs for Employee Compensation, release USDL-26-0827, reference March 2026, private industry.","unit":"x wages, for total employer cost","amountUsd":1.43,"note":"Wages are 69.9% of employer cost and benefits 30.1%. A $120,230 salary costs roughly $172,000 to employ.","source":"https://www.bls.gov/news.release/pdf/ecec.pdf","verified":"2026-08-27"},{"layer":"labor","vendor":"Levels.fyi","product":"Machine Learning Engineer","measure":"median total compensation","basis":"Self-reported, US, weighted toward large technology employers.","unit":"per year","amountUsd":279000,"note":"Not comparable to a BLS wage, because it includes equity. The same site on the same day puts \"AI Engineer\" at $154,000: an 81% gap between two titles for similar work.","source":"https://www.levels.fyi/t/software-engineer/title/machine-learning-engineer","verified":"2026-08-27"},{"layer":"labor","vendor":"YunoJuno","product":"AI Engineer, freelance","measure":"average contract rate","basis":"Actual bookings, from more than 182,000 contractor and booking data points across 2024 and 2025.","unit":"per hour","amountUsd":70,"note":"Machine Learning Engineer books at $62. The report is titled 2026 but runs on 2024 to 2025 data.","source":"https://www.yunojuno.com/freelancer-rates-report","verified":"2026-08-27"},{"layer":"labor","vendor":"Arc.dev","product":"AI and ML engineering, freelance","measure":"average SENIOR contract rate","basis":"Survey of 5,302 freelance developers worldwide, dated 2026-05-26.","unit":"per hour, low end of range","amountUsd":110,"note":"The range runs to $190. Read the column header: this is senior-only, not a market midpoint, which is most of why it sits about 2x above booking data.","source":"https://arc.dev/employer-blog/freelance-developers-cost/","verified":"2026-08-27"},{"layer":"labor","vendor":"Clutch","product":"AI development firms, US and Canada","measure":"agency hourly band","basis":"Directory of firms with verified client reviews.","unit":"per hour, low end of band","amountUsd":50,"note":"The band runs to $99 and the modal global rate is $24 to $49. The same directory shows an identical band for general software development.","source":"https://clutch.co/developers/artificial-intelligence/pricing","verified":"2026-08-27"},{"layer":"labor","vendor":"Clutch","product":"Average AI project","measure":"project total","basis":"Directory of firms with verified client reviews.","unit":"per project","amountUsd":120594.55,"note":"The most common range is $10,000 to $49,999. The comparable figure for general software projects is $132,480.29, which is HIGHER. At directory level there is no AI premium.","source":"https://clutch.co/developers/pricing","verified":"2026-08-27"}],"scheduledChanges":[{"vendor":"Anthropic","what":"Claude Sonnet 5 stays at $2 / $10 per 1M. The rise to $3 / $15 that was scheduled for 2026-09-01 has been cancelled and the launch rate is now standard.","effective":"2026-09-01 (cancelled)","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"},{"vendor":"OpenAI","what":"gpt-5.6-sol is on promotional pricing, stated as available at least through 2026-11-21.","effective":"2026-11-21","source":"https://developers.openai.com/api/docs/pricing","verified":"2026-08-27"},{"vendor":"Google","what":"Gemini 3.7 and 3.6 Flash roughly double: input $0.75 to $1.50, output $3.75 to $7.50, with caching and storage rising in step.","effective":"2027-01-01","source":"https://ai.google.dev/gemini-api/docs/pricing","verified":"2026-08-27"},{"vendor":"Anthropic","what":"Claude 4.7 and later use a newer tokenizer that produces roughly 30% more tokens for the same text. Per-token prices did not change, but cost per task is not comparable across the 4.6 / 4.7 line.","effective":"in force","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"}],"researchFindings":[{"org":"BCG","finding":"In AI work, algorithms are about 10% of the effort, technology about 20%, and people, process and organisation about 70%. Among those surveyed, 74% named a shortage of AI talent as a constraint and 63% said scaling costs were hard to control.","sample":"1,250 senior executives across 68 countries","published":"September 2025","source":"https://media-publications.bcg.com/The-Widening-AI-Value-Gap-Sept-2025.pdf","verified":"2026-08-27"},{"org":"IBM Institute for Business Value","finding":"Only 25% of AI initiatives have delivered the return that was expected of them, and only 16% have scaled across the enterprise.","sample":"2,000 CEOs across 33 countries, fielded February to April 2025","published":"May 2025","source":"https://newsroom.ibm.com/2025-05-06-ibm-study-ceos-double-down-on-ai-while-navigating-enterprise-hurdles","verified":"2026-08-27"},{"org":"Gartner","finding":"Over 40% of agentic AI projects will be cancelled by the end of 2027.","sample":"Poll of 3,412 webinar attendees, January 2025","published":"June 2025","caveat":"This is a webinar poll, not a controlled survey. Read it as a signal about expectations, not a measured failure rate.","source":"https://www.gartner.com/en/newsroom/press-releases/2025-06-25-gartner-predicts-over-40-percent-of-agentic-ai-projects-will-be-canceled-by-end-of-2027","verified":"2026-08-27"},{"org":"MIT NANDA","finding":"Widely quoted as finding that 95% of organisations get zero return from generative AI. The same report found externally built systems reached deployment about twice as often as internally built ones.","sample":"300+ initiatives, 52 organisations interviewed, 153 leaders surveyed","published":"July 2025","caveat":"Contested, and included mainly so the caveat travels with the number. The document is labelled preliminary findings, is not peer reviewed, and its own limitations section concedes the build-versus-buy split rests on interviews rather than market data. The methodology repeated in the press, \"150 interviews and 350 employee surveys\", is a misquote of the figures above.","source":"https://cloudelligent.com/wp-content/uploads/2026/02/v0.1_State_of_AI_in_Business_2025_Report.pdf","verified":"2026-08-27"}],"vendorWorkedExamples":[{"vendor":"Anthropic","scenario":"10,000 customer-support conversations","detail":"About 3,700 tokens per conversation on Claude Haiku 4.5 at $1/$5 per 1M.","totalUsd":37,"perUnitUsd":0.0037,"unitLabel":"per conversation","source":"https://platform.claude.com/docs/en/about-claude/pricing","verified":"2026-08-27"}],"fiveMinuteCall":[{"vendor":"Retell AI","configuration":"Their own default calculator, bring-your-own SIP","fiveMinuteUsd":0.55,"resolvable":true,"breakdown":"Voice infrastructure $0.055 + model $0.04 + speech synthesis $0.015 per minute, telephony $0 on your own SIP. $0.11 per minute.","source":"https://www.retellai.com/pricing","verified":"2026-08-27"},{"vendor":"Retell AI","configuration":"Premium voices and a top-tier model","fiveMinuteUsd":1.35,"resolvable":true,"breakdown":"Voice infrastructure $0.055 + ElevenLabs synthesis $0.040 + GPT-5.5 $0.16 + Twilio $0.015. $0.27 per minute, or roughly 2.5x the default.","source":"https://www.retellai.com/pricing","verified":"2026-08-27"},{"vendor":"Bland AI","configuration":"Start plan, Twilio local inbound","fiveMinuteUsd":0.74,"resolvable":true,"breakdown":"$0.14 per minute covering model, speech-to-text and synthesis, plus about $0.04 of telephony across five minutes.","source":"https://www.bland.ai/pricing","verified":"2026-08-27"},{"vendor":"ElevenLabs","configuration":"Agents, additional minutes","fiveMinuteUsd":0.4,"resolvable":false,"breakdown":"$0.08 per minute for the platform. The model and telephony are billed at cost on top, so the total depends on which model you attach.","source":"https://elevenlabs.io/pricing/agents","verified":"2026-08-27"},{"vendor":"Vapi","configuration":"Build plan, components at cost","fiveMinuteUsd":0.32,"resolvable":false,"breakdown":"A verifiable FLOOR, not a price: hosting $0.25 + speech-to-text $0.024 + telephony $0.043. Speech synthesis is per character and the model is per token, so neither can be resolved without knowing how much the agent says.","source":"https://vapi.ai/pricing","verified":"2026-08-27"}]}