Lyzr AI Pricing 2026: The Per-Agent-Run Model

TL;DR: Lyzr AI pricing is usage-based at $0.08 per agent run on Lyzr Cloud and $0.03 per agent run self-hosted (VPC or on-prem), with no per-seat fees. The catch: LLM tokens are billed separately as a pass-through cost, and a real workflow can use many agent runs, so a single completed task (like a KYC check) can cost around $1, not $0.03.

Last updated: August 2026 · By Sandeep Singh, DailyFix

Lyzr AI pricing in 2026: cheap per run, but LLM tokens are separate and real workflows chain many runs.

Lyzr is an enterprise AI agent platform that got attention in 2026 partly by letting one of its own agents help run a $100M fundraise. Its pricing pitch is refreshingly concrete: one price per agent run, benchmarked against human labor. But two details, what a run includes and what it excludes, decide your real bill. Here is the math, checked in August 2026.

How does Lyzr AI pricing work in 2026?

Lyzr AI pricing is usage-based and charged per agent run, with no seats or lock-in. As of August 2026, Lyzr Cloud (fully managed) is $0.08 per agent run and Lyzr VPC / On-Prem (self-hosted) is $0.03 per agent run, per Lyzr's pricing page. Enterprise terms are arranged through a sales conversation.

Instead of monthly seats, you pay for what actually executes. That model rewards efficient agents and punishes nothing you do not run. The trade-off is that the flat per-run number is only part of the cost picture: language-model tokens sit outside it, and on the self-hosted tier so does compute. Understanding those two exclusions is the difference between the $0.03 sticker and your invoice.

What is an agent run in Lyzr?

An agent run in Lyzr is one full execution of an agent, and the single run price bundles the entire platform stack: the knowledge base call, tool call, agent call, memory call, Responsible AI guardrails, and agent security policy, per Lyzr's pricing page checked in August 2026. One charge covers all of that internal work.

That bundling is genuinely good value, because a run that touches memory, calls a tool, and passes through guardrails still costs one flat run price rather than metering each sub-step. The catch is scope: an "agent run" is one agent executing once. A useful business process usually orchestrates many agents, so your cost scales with how many runs the workflow chains together, not with the single-run price you first read.

Option	Price / agent run	Hosting	LLM cost	Compute cost
Lyzr Cloud	$0.08	Fully managed	Separate (pass-through)	Included
Lyzr VPC / On-Prem	$0.03	Self-hosted	Separate (pass-through)	Your usage
Enterprise	Custom	Cloud or VPC	Separate	Custom

Prices and limits change often — verify on the official pricing page before deciding.

How much does a real Lyzr workflow cost?

A real Lyzr workflow costs far more than one run price, because most processes chain many agents. Lyzr's own example, a KYC check for corporate banking customers, needs 20-plus agents and works out to about $1.02 per completed KYC run at the $0.03 self-hosted rate, per Lyzr's pricing page checked in August 2026.

That example is the honest way to budget Lyzr: think in cost-per-completed-task, not cost-per-run. If one business outcome takes 20 to 34 agent runs, multiply the per-run price accordingly, then add the separate LLM tokens each of those agents consumed. The upside Lyzr emphasizes is that roughly $1 to automate a full KYC still undercuts manual processing by a large margin, which is the real selling point. But you cannot forecast spend from the $0.03 figure alone; you forecast it from runs-per-task times tasks-per-month, plus model usage.

Does Lyzr include LLM costs?

No. Lyzr's per-agent-run price covers the platform stack only; LLM costs are billed separately at pass-through usage rates, and on VPC or on-prem your compute is usage-based too, per Lyzr's pricing page checked in August 2026. The run fee is not an all-in number.

This is the exclusion most people miss when they see "$0.03 per agent run." Every agent that reasons calls a language model, and those tokens land on a separate line at the provider's rates. For a token-heavy agent working over long documents, the LLM cost can rival or exceed the Lyzr run fee itself. The practical takeaway: when you compare Lyzr to a flat-priced competitor, add expected model tokens to Lyzr's run price so you are comparing all-in numbers, not a platform fee against a bundled one.

Is Lyzr worth it for AI agents?

Lyzr for AI agents is worth it when you need governed, production-grade agents with compliance built in, SOC 2, HIPAA, GDPR, ISO 27001, guardrails, and a control plane, rather than a quick hobby bot. The usage-based per-run model can make automated work dramatically cheaper than manual labor once workflows are live.

The honest fit test is who you are. Lyzr's guided onboarding, self-hosting option, and separate LLM and compute costs point clearly at teams with real, repeatable workflows and someone to own deployment, not a solopreneur wanting an instant self-serve bot. If you run high-volume, structured processes like onboarding, claims, or KYC, Lyzr's per-run economics are compelling and transparent. If you just want to experiment cheaply this weekend, a self-serve agent builder will get you moving faster.

Frequently asked questions

How much does Lyzr cost per agent run?
Lyzr costs $0.08 per agent run on Lyzr Cloud, its fully managed option, and $0.03 per agent run on Lyzr VPC or On-Prem, where you self-host, per Lyzr's pricing page checked in August 2026. There are no per-seat fees. LLM usage is billed separately as a pass-through cost on top of the run price.

What is an agent run in Lyzr?
An agent run in Lyzr is one full execution of an agent, and the single run price bundles the whole stack: the knowledge base call, tool call, agent call, memory call, Responsible AI guardrails, and agent security policy. So one flat run charge covers everything the agent does internally, except the language model tokens, which are billed separately.

Does Lyzr pricing include LLM costs?
No. Lyzr's per-agent-run price covers the platform stack, but LLM costs are billed separately at pass-through usage rates, per Lyzr's pricing page checked in August 2026. On VPC or on-prem, compute costs are also usage-based. So your real bill is the run price plus model tokens plus, if self-hosted, your own compute.

Does Lyzr have a free or self-serve plan?
Lyzr publishes flat per-agent-run prices, $0.08 cloud and $0.03 self-hosted, but the pricing page routes every plan through a Talk to Us step rather than an instant self-serve checkout. You can explore Agent Studio to build agents, and Lyzr positions an 8-week path from use case to production with its team, so onboarding is guided rather than pure self-serve.

How much does a real Lyzr workflow cost?
It depends on how many agent runs a workflow needs. Lyzr's own example, a KYC process for corporate banking, uses 20-plus agents and works out to about $1.02 per completed KYC run at the $0.03 self-hosted rate. So the meaningful number is cost-per-completed-task, not the headline per-run price, plus separate LLM tokens.

Is Lyzr good for building AI agents?
Lyzr is aimed at enterprises that want governed, production AI agents, with SOC 2, HIPAA, GDPR, ISO 27001, guardrails, and a control plane. The usage-based per-run model can be very cheap per task, but the guided onboarding and separate LLM and compute costs make it better suited to teams with real workflows than to a solo hobbyist.

Bottom Line

Lyzr is one of the clearest-priced enterprise AI agent platforms in 2026, at $0.08 per agent run on cloud or $0.03 self-hosted, with governance and compliance built in. Budget it in cost-per-completed-task, because real workflows chain many runs, and always add the separate LLM and, if self-hosted, compute costs. Choose Lyzr if you run structured, high-volume processes and have someone to own deployment. If you want an instant, free self-serve bot to tinker with, look elsewhere first.

Sources

Lyzr — Official pricing (checked August 2026)
Lyzr — Documentation

About the author — Sandeep Singh. Sandeep builds and tests AI tools and automations daily, and runs DailyFix to show solopreneurs which ones actually save time.

Sandeep Singh Founder & Editor, DailyFix

Sandeep Singh is the founder and editor of DailyFix. He served in the Border Security Force for five years before moving into software, where he now builds and tests AI automations for solopreneurs and small teams — mostly in n8n, Make, and Zapier, plus voice agents on Vapi and Synthflow. He runs DailyFix on the same principle he uses in his own work: no tool goes into a post unless he'd actually put it into a client build. Every review, prompt and workflow on the site is tested end-to-end on real projects before it's written up. Alongside DailyFix, Sandeep ships SwiftSite Studio. Recent focus: AI receptionists for contractors, and turning ranking-audit output into repeatable, one-person publishing systems. Reach him on GitHub.

#Lyzr AI pricing #Lyzr agent run #AI agent platform #Lyzr for AI agents #enterprise AI agents #Lyzr cost