Lyzr Pricing in 2026: How Much Does Lyzr Really Cost

Lyzr pricing in 2026 combines a tiered subscription model for build-time access with usage-based billing for production agent runs. Subscriptions start with a free Community plan, move through a Starter plan at $19 per month, a Pro plan at $99 per month (or $79 per month billed annually), and an Enterprise tier that is custom-quoted, according to Lyzr's public pricing page and the official Lyzr documentation. Production workloads are priced at $0.08 per agent run on Lyzr Cloud and $0.03 per agent run on Lyzr VPC or On-Premise deployments, with large language model (LLM) costs passed through separately at market rates.

This guide breaks down how much Lyzr really costs in the 2026  plan by plan, by team size, including additional costs most buyers underestimate, a full three-year total cost of ownership (TCO) model, and how to negotiate enterprise contracts. It is written for AI platform engineers, heads of automation, RevOps leaders, and procurement teams evaluating agentic AI infrastructure in 2026.

Key Takeaways

• Lyzr has published transparent pricing for its lower tiers: free Community, $19 Starter, and $99 Pro plans are listed publicly, which is notable in the otherwise quote-heavy enterprise AI agent category.
• Enterprise pricing is custom-quoted: Unlimited credits, 50+ builder licenses, 120 GB knowledge base, 24/7 support, and a 48-hour custom integration SLA sit behind a "Contact Sales" button.
• Production runs are billed per agent run: $0.08 per agent run on Lyzr Cloud and $0.03 per agent run on Lyzr VPC or on-prem.
• LLM tokens are pass-through: at transparent market rates and are not included in the credit bundles. Plan for separate OpenAI, Anthropic, Google, or Bedrock invoices alongside your Lyzr bill.
• Extra credit top-ups: run from $10 for 1,000 credits up to $5,000 for 500,000 credits, offering a way to smooth usage spikes without switching tiers.
• Typical TCO for a 5,000-employee organization: running Lyzr Enterprise on a moderate workload commonly lands between $220,000 and $650,000 in year one, once platform, production runs, LLM spend, and internal time are included.
• Multi-year commitments meaningfully lower effective cost: Lyzr already discounts Pro by roughly 20 percent for annual billing, and Enterprise deals have further room through standard procurement levers.

What Lyzr Sells in 2026

Lyzr is a full-stack agentic AI framework for enterprises  the infrastructure layer that teams use to design, deploy, govern, and run autonomous AI agents in production. The core agentic framework is open source, and Lyzr markets the platform on data sovereignty, built-in Responsible AI guardrails, and the ability to deploy to Lyzr Cloud, a customer VPC, or a fully on-premise environment. Lyzr's $8M Series A announcement positions the company as building an "Agentic Operating System" for the enterprise, with follow-on reporting from TechFront360 noting a $14.5M cumulative raise as enterprise AI agents became core infrastructure through 2026.

On top of the framework, Lyzr ships a catalog of pre-built "Super Agents"  vertical agents designed for specific business functions:

• Jazon is marketed as "the world's first truly agentic AI SDR." Jazon handles account research, ICP scoring, email discovery, personalized outreach, and automated follow-up. The vendor lists a 24-hour deployment path through the low-code framework.
• Skott is an autonomous marketing agent that automates content, campaigns, SEO, and analytics across a full marketing funnel.
• Diane is a generic operational agent template that many customers use as a starting point for internal support, operations, or knowledge workflows.
• Chat and knowledge agents: retrieval-augmented conversational agents that sit on top of enterprise knowledge bases for customer support, employee support, or sales enablement.

Lyzr positions itself as an enterprise alternative to open-source frameworks like CrewAI, AutoGen, and LangGraph, adding production-grade governance, security, and managed infrastructure. Lyzr's own comparison of open-source agentic frameworks describes the platform as "built to natively handle components from the top open source agentic frameworks,"  a positioning that appeals to engineering teams who want the flexibility of open source with the operational hardening of a commercial platform.

Why Lyzr Pricing Matters Now in 2026

Three forces make 2026 an important year to understand agentic AI pricing models carefully, and Lyzr pricing in particular.

First, enterprise AI budgets are under fresh scrutiny.

 After two years of aggressive AI adoption, CFOs and CIOs are asking harder questions about unit economics. According to Lyzr's 2026 State of AI Agents report, deployment of production AI agents has moved from pilot to core infrastructure, which means pricing now lives in the capital-planning cycle rather than the innovation-pilot line.

Second, the agentic AI category is consolidating around hybrid pricing models.

 The historical split between flat-rate subscriptions and consumption-based billing is resolving into a hybrid: you subscribe for build-time access (licenses, storage, support) and you pay per run for production. Lyzr's model subscription, plus $0.08 or $0.03 per agent run  is one of the cleaner public examples of this hybrid structure.

Third, LLM pass-through economics matter more than platform fees at scale. 

For any organization running thousands or millions of agent invocations per month, token costs to OpenAI, Anthropic, Google, or a self-hosted model frequently dominate the bill. The platform fee is the floor; the real cost curve is the LLM line.

A disciplined Lyzr evaluation in 2026, therefore, needs to answer three questions at once: what does the platform cost, what does production usage cost, and what do the underlying model tokens cost? This guide tackles all three. For a broader context on how agentic AI economics are reshaping enterprise buying behavior, see Landbase's analysis of agentic AI statistics.

Lyzr Plan Breakdown: Community, Starter, Pro, Enterprise

Lyzr publishes subscription tiers for build-time access. Every tier includes unlimited agents and unlimited users on the paid plans; what changes tier-to-tier is credits, storage, log retention, number of builder licenses, and support depth.

Community (Free)

The free Community plan is designed for prototyping, learning, and individual developer use.

• Price: $0
• Credits: 500 per month
• Builder licenses: 1
• Knowledge base: 100 MB
• Agents: Up to 10 (per the Lyzr documentation)
• Log retention: 7 days or none, depending on the specific feature
• Models: Base model set
• Support: Email or community

Who it fits: Solo developers, AI engineers prototyping agent flows, early-stage startups sanity-checking an approach before committing to paid tooling.

What's missing: Log retention depth, premium models, and the operational governance features that matter for production deployments.

Starter ($19/month)

The Starter tier is the lightest paid plan and sits between hobbyist use and small-team production workloads.

• Price: $19 per month
• Credits: 2,000 per month
• Builder licenses: 1
• Knowledge base: 100 MB
• Agents: Up to 15
• Log retention: 7 days
• Support: Email

Who it fits: Small teams running low-volume agents against a focused knowledge base  customer-facing chatbots for a single product line, internal knowledge assistants for a small operations team, or a proof-of-value deployment ahead of an enterprise commitment.

Pro ($99/month or $79/month billed yearly)

Pro is Lyzr's most popular self-serve tier and is where many mid-sized teams run their first production workload.

• Price: $99 per month (monthly billing) or $79 per month billed annually  a 20 percent discount that works out to $948 per year prepaid
• Credits: 10,000 per month (120,000 on the annual plan)
• Builder licenses: 1
• Knowledge base: 1 GB
• Agents: Up to 25
• Log retention: 3 months on the listed feature set; some configurations show 7 days
• Support: Email with limited Super Agents support
• Models: All standard models

Who it fits: Growing startups, small AI engineering teams, and internal innovation groups running a handful of agents in staging or low-volume production. Pro is the tier where most first-time buyers discover whether Lyzr fits their workflow before moving to Enterprise.

Enterprise Cloud (Custom Pricing)

Enterprise is where Lyzr unlocks the features required for production-grade, governance-heavy deployments across large organizations.

• Price: Custom  contact sales
• Credits: Unlimited
• Builder licenses: 50 or more
• Knowledge base: 120 GB
• Log retention: 1 year
• Models: Standard, premium, and custom (bring-your-own-model)
• Super Agents: Up to 10
• Support: 24/7 technical support
• Integration SLA: 48-hour custom integration SLA
• Governance: Agent Entitlement Policy, Organizational General Intelligence
• Professional services: Lyzr Build Services for 5 agents included

Who it fits: Large enterprises operating agents across multiple business units, regulated industries that need audit trails and responsible AI documentation, and platform engineering teams that want 24/7 support and named SLAs.

On-Premise (Custom Pricing)

For organizations that cannot or will not send data to a vendor cloud, Lyzr offers an on-premise deployment.

• Price: Custom
• Credits: Unlimited
• Log retention: Retained forever within your environment
• Super Agents: Up to 20
• Integration SLA: Custom
• Compute: Usage-based additional

Who it fits: Financial services firms, healthcare systems, government contractors, and any organization with strict data residency or air-gapped requirements.

Extra Credits

All tiers allow one-time credit top-ups for usage spikes. Public pricing runs from $10 for 1,000 credits up to $5,000 for 500,000 credits, with standard volume discounts along the curve. This is useful for teams on Pro or Starter that hit a seasonal spike without warranting an Enterprise upgrade.

Cost by Team Size

How Lyzr pricing shapes up depends heavily on how many agent builders you have and how much production traffic you expect. Below is how the economics commonly play out by team size.

Solo Developer or Evaluator (1 user)

• Likely plan: Community (free) for initial exploration, Starter ($19/mo) or Pro ($99/mo) once prototypes stabilize.
• Platform cost year 1: $0 to $948
• Production run cost: Minimal  typically absorbed by credits unless volume exceeds the tier bundle.
• What to budget: A few hundred dollars of LLM tokens during experimentation.

Small Team (2 to 5 users)

• Likely plan: Multiple Pro seats ($79 to $99/user/month) or a light Enterprise starter agreement.
• Platform cost year 1: $1,900 to $6,000 on Pro seats, or a negotiated Enterprise starter around $15,000 to $40,000 depending on scope.
• Production run cost: Depends entirely on workload  50,000 agent runs per month on Lyzr Cloud works out to $4,000 per month, or $1,500 per month on VPC.
• What to budget: Platform plus production runs plus moderate LLM spend, typically $25,000 to $75,000 for year one.

Mid-Market (5 to 25 users)

• Likely plan: Enterprise Cloud with 10 to 25 builder licenses, higher credit caps, and some Super Agent deployment services.
• Platform cost year 1: Typically $60,000 to $200,000 based on industry benchmarks for enterprise AI agent platforms in 2026.
• Production run cost: Organizations in this band commonly run 500,000 to 5 million agent invocations per month, translating into $15,000 to $300,000 annually on Lyzr Cloud at list rates, materially lower on VPC.
• LLM spend: Frequently equal to or larger than the platform line. Budget $50,000 to $250,000 annually.

Enterprise (25+ users, large production volume)

• Likely plan: Enterprise Cloud with a negotiated credit pool, or On-Premise for regulated workloads.
• Platform cost year 1: Commonly $150,000 to $600,000 based on publicly available benchmarks for agentic AI platforms in this band.
• Production run cost: Highly variable. A deployment running 20 million agent runs per year on VPC pricing lands at roughly $600,000 at list.
• LLM spend: Frequently the dominant cost line, particularly if premium frontier models are in the agent path.

The consistent pattern is that platform fees rarely exceed 30 to 40 percent of year-one total cost once production runs, LLM tokens, and internal staff time are included. Budget accordingly.

Additional Costs to Plan For

Sticker price is only part of the conversation. The following cost categories are routinely under-budgeted by teams running their first agentic AI deployment.

• LLM Token Costs (Pass-Through): Lyzr's pricing page explicitly notes that "LLM costs are billed separately at transparent, pass-through usage rates." Every agent call that invokes a foundation model incurs a token bill to the underlying provider  OpenAI, Anthropic, Google Gemini, AWS Bedrock, or a self-hosted open-weights model. For reasoning-heavy agents or long-context workflows, token spend routinely exceeds platform spend at scale.
• Compute Costs (On-Premise Deployments): For Lyzr On-Premise, the pricing page notes that "compute costs are also usage-based" on top of the subscription. Whether you run on your own GPU cluster, a private cloud instance, or a dedicated Kubernetes environment, the compute line is yours to manage. Factor in GPU time, container orchestration overhead, storage, and observability tooling, typically a six-figure capex or opex line at enterprise scale.
• Implementation and Integration Services: Enterprise Lyzr contracts often include Lyzr Build Services for a defined number of agents. The Enterprise Cloud tier lists five agents of build services. Beyond that allotment, organizations engage Lyzr Professional Services or third-party implementation partners to build additional agents, integrate with line-of-business systems, and tune the knowledge base.
• Knowledge Base Preparation: Agentic AI quality is a function of the source documentation the agents can read. Teams with fragmented, stale, or inconsistent knowledge bases commonly spend $20,000 to $75,000 in either internal time or external services cleaning up Confluence, SharePoint, Google Drive, and product wiki content before agent accuracy reaches acceptable thresholds.
• Internal Staff Time: A typical mid-market Lyzr deployment usually needs an AI platform engineer or technical owner (0.5–1.0 FTE for the first three months), an integration engineer across IT/data/security (0.25 FTE for three months), a knowledge admin (0.1–0.25 FTE ongoing), and an executive sponsor plus a governance reviewer. Overall, internal staff time often costs between $50,000 and $120,000 in year one, and it won’t appear on any vendor invoice.
• Monitoring and Observability: Production agentic AI deployments commonly add a dedicated observability stack on top of Lyzr's built-in logs, tools like LangSmith, Arize, Weights & Biases, or a custom solution. Budget $15,000 to $60,000 per year for observability tooling, depending on breadth.
• Security and Compliance Review: Enterprise agentic AI deployments typically trigger a security review, a privacy impact assessment, and, in regulated industries, a model risk management review. Budget internal security and legal time, frequently $20,000 to $75,000 in loaded cost for the initial review, plus annual renewal.

What's Included in Each Plan

Lyzr's platform includes a common core across all tiers, with depth and governance features unlocking at higher tiers.

Core Platform (All Tiers)

• Agent framework: Open-source core, used to design single-agent and multi-agent workflows.
• Knowledge base: Vector storage and retrieval with configurable chunking and embedding strategies.
• Tool calling: Native support for custom Python tools, API integrations, and common enterprise connectors.
• Memory: Short-term and long-term memory primitives for persistent agent state.
• Responsible AI guardrails: Content filters, PII redaction, and safety policies applied per agent.
• Agent Security Policy: Role-based access controls on what agents can read, write, and execute.

Pro-Level Additions

• Increased credits and storage (10,000 credits per month, 1 GB knowledge base)
• All standard models are accessible from the agent builder
• Limited Super Agents support  meaning access to a subset of pre-built agents
• 3-month log retention for debugging and audit review

Enterprise-Level Additions

• Unlimited credits: moving production economics entirely to the usage-based agent run rate
• 50+ builder licenses: supporting platform teams across business units
• 120 GB knowledge base: enterprise-scale retrieval
• Premium and custom models: including bring-your-own-model for in-house fine-tunes or proprietary LLMs
• Organizational General Intelligence: Lyzr's term for a cross-agent shared reasoning layer
• Agent Entitlement Policy: fine-grained permissioning across agents and users
• 10 Super Agents: full catalog access including Jazon, Skott, Diane, and others
• 48-hour custom integration SLA: contractually bound turnaround on enterprise connector builds
• 24/7 support and named customer success
• Lyzr Build Services for 5 agents: professional services allotment for initial deployment

On-Premise-Specific Additions

• Infinite log retention: logs stay entirely within the customer environment
• 20 Super Agents: the widest catalog access
• Custom integration SLA: negotiated to customer requirements
• Full data sovereignty: no data leaves the customer environment at any point

Illustrative Three-Year Total Cost of Ownership Example

To show how list pricing translates into real cost, here is a sample three-year TCO model for a hypothetical mid-market enterprise running Lyzr Enterprise Cloud with a moderate agent workload  eight production agents, 3 million agent runs per year, and premium LLM usage. All figures are illustrative and based on publicly available list pricing; actual quotes and production profiles will vary.

Assumptions

• 15 builder licenses under the Enterprise plan
• Negotiated an annual platform fee of $120,000
• 3 million agent runs per year on Lyzr Cloud at $0.08 per run = $240,000 per year
• LLM token spend of $180,000 per year (mix of frontier and standard models)
• Implementation services: $75,000 one-time
• Knowledge base preparation: $35,000 one-time
• Ongoing professional services: $25,000 per year
• Internal staff time: $90,000 in year 1, $50,000 ongoing
• Observability tooling: $30,000 per year
• Renewal uplift: 6% capped (negotiated)

Year 1 Cost Breakdown

• Annual platform license: $120,000. 
• Production agent runs (3M at $0.08): $240,000. 
• LLM token spend: $180,000. 
• Implementation services: $75,000. 
• Knowledge base preparation: $35,000. 
• Ongoing professional services: $25,000. 
• Internal staff time: $90,000. 
• Observability tooling: $30,000. 
• Year 1 Total: $795,000.

Three-Year TCO

• Year 1: Platform, runs, and LLM: $540,000. Services and internal time: $255,000. Annual total: $795,000.
• Year 2: Platform, runs, and LLM: $572,400. Services and internal time: $105,000. Annual total: $677,400.
• Year 3: Platform, runs, and LLM: $606,744. Services and internal time: $105,000. Annual total: $711,744.

Three-year TCO: $2,184,144 ($1,719,144 in platform, runs, and LLM; $465,000 in services and internal time).

What This Tells You

Platform fees represent roughly 17% of three-year TCO; agent runs and LLM tokens dominate the model. Moving to VPC ($0.03 per run) would reduce the agent run line by over 60%, representing a swing of approximately $450,000 over three years, and routing routine agent calls to smaller or self-hosted models while reserving frontier models for reasoning-heavy steps can cut LLM spend by 40–60% without meaningful quality loss. 

Year-one costs are front-loaded due to implementation and knowledge base preparation; steady-state years 2 and 3 run roughly 15% lighter.

How to Get the Best Deal on Lyzr

Lyzr's self-serve tiers have limited negotiation room; the list price is the price. Enterprise deals, however, have meaningful flexibility across several standard procurement levers.

• Commit Annually on Paid Tiers: The Pro plan's annual billing discount of $79 per month prepaid versus $99 monthly is effectively a 20 percent discount for locking in 12 months. For any team expecting to use Lyzr beyond a single quarter, annual billing is the cleanest immediate saving.
• Negotiate Blended Agent Run Rates: Enterprise contracts typically blend platform subscription with a committed volume of agent runs. Rather than paying a list price of $0.08 per run on Lyzr Cloud, procurement teams commonly negotiate tiered rates, for example, $0.06 per run up to a committed monthly volume, with overages at list. A committed volume of 2 million or more runs per month is the threshold where blended rates typically become available.
• Consider VPC or On-Premise for High-Volume Workloads: The published $0.03 per run rate on Lyzr VPC and on-prem is more than 60 percent cheaper than Lyzr Cloud's $0.08. For any organization running more than 1 million agent runs per month, the savings over three years routinely exceed the added infrastructure and operational cost of managing a VPC or on-prem deployment.
• Request Build Services as Part of the Enterprise Package: The Enterprise Cloud tier lists Lyzr Build Services for five agents included. For deployments that will exceed five initial agents, negotiate additional build services hours into the initial contract rather than paying at list for professional services later.
• Run a Genuine Competitive Process: A structured evaluation with at least two other agentic AI platforms in the RFP meaningfully improves negotiation outcomes. Document the alternatives in writing, procurement teams across enterprise SaaS consistently report that concrete competitive risk drives deeper discounts than soft pressure.
• Time the Deal to Quarter-End: Like most enterprise software vendors, Lyzr's sales cycles are quota-driven. Quarter-end and year-end quota pressure consistently deliver better pricing on enterprise agreements. Plan your signature timeline backwards from the relevant quarter-end.
• Negotiate Renewal Caps in the Initial Deal: Insist on renewal cap language in the initial contract, commonly 5 to 7 percent maximum annual uplift, with a right to true-down committed agent run volumes if usage runs below plan. Without caps, renewal uplifts in the agentic AI category have historically run 10 to 25 percent.
• Ask About Early-Customer or Case Study Pricing: Lyzr, like most growth-stage enterprise AI vendors, periodically offers discounted pricing in exchange for case study rights, conference speaking slots, or reference customer availability. If your brand carries weight, ask.

Tools and Solutions for Enterprise AI Agents

Several categories of tools address overlapping problems in the agentic AI space. Buyers evaluating Lyzr commonly also consider:

Landbase — Agentic AI Go-to-Market Platform

What They Do: Landbase is an AI-native GTM platform that automates the end-to-end work of targeting, qualifying, and engaging accounts through agentic, multi-agent workflows. The platform combines a 300M+ verified B2B contact database, 1,500+ enrichment and signal fields, AI qualification, and multi-channel outreach execution in a single system — without requiring separate prospecting, sequencing, or analytics tools.

Why They're Important: Landbase is the only platform that takes a team from a natural-language prompt to a verified, scored account list to a launched multi-channel campaign without switching tools. GTM-2 Omni, trained on 40M+ B2B campaigns and 175M+ sales conversations, continuously qualifies and prioritizes accounts across the full TAM. Customers report 4–7x higher conversion rates, with outcomes including $400K MRR added by P2 Telecom and 33% more meetings booked by Digo Media without additional headcount. Gartner recognized Landbase as a Cool Vendor 2025 in AI-driven GTM automation.

Key Stats / Metrics:

• 300M+ verified B2B contacts
• 24M+ companies with 1,500+ enrichment fields
• 97%+ database deliverability
• $30M Series A raised in January 2026

Leadership: CEO: Daniel Saks (co-founder, formerly co-CEO of AppDirect, Forbes 30 Under 30)

Founded: 2020

Recent Funding: Series A: $30M (January 2026) Investors: Picus Capital, 8VC, Sound Ventures

Open-Source Agentic Frameworks (CrewAI, AutoGen, LangGraph)

What They Do: Open-source frameworks provide maximum flexibility for engineering teams that want to own the full agent stack. There are no platform fees, and the frameworks are highly customizable for specific enterprise use cases.

Why They're Important: These frameworks offer the broadest degree of architectural control and have active open-source communities. Production hardening, security, governance, and managed infrastructure are the responsibility of the engineering team, making them best suited for organizations with dedicated AI platform capacity.

Key Stats / Metrics:

• Zero platform license cost
• Community-driven development and support

Leadership: Community-maintained (varies by project)

Founded: 2023–2024 (varies by framework)

Hyperscaler Agent Platforms (AWS Bedrock AgentCore, Vertex AI Agent Builder, Azure AI Foundry)

What They Do: Hyperscaler platforms provide native agent-building capabilities deeply integrated with a cloud provider's data, identity, and infrastructure stack. Pricing is tied to the hyperscaler's broader contract and consumption model.

Why They're Important: These platforms offer deep integration with existing cloud environments and identity systems, making them well-suited for organizations already standardized on a single cloud provider. Governance, security, and observability are handled within the existing cloud management plane.

Key Stats / Metrics:

• Usage-based pricing tied to cloud consumption
• Native integration with provider data and identity services

Leadership: AWS (Amazon), Google Cloud, Microsoft Azure

Founded: Launched 2023–2025 (varies by platform)

How Landbase Redefines the GTM Stack

For go-to-market teams rethinking their outbound motion from the ground up, Landbase was purpose-built to address the core gap that agentic AI frameworks leave unfilled: the intelligence, qualification, and activation work that GTM teams need done, without requiring a dedicated engineering team to build and govern custom agents.

With Landbase, teams describe their ideal customer in plain English and receive a verified, AI-qualified audience list enriched across 1,500+ signal fields, ready for CRM sync and immediate campaign activation. GTM-2 Omni, trained on 40M+ B2B campaigns, handles targeting, scoring, enrichment, and multi-channel outreach coordination autonomously, reducing manual research effort by approximately 80%.

Where a framework-based agentic AI platform like Lyzr is designed for engineering teams building horizontal agent infrastructure across the enterprise, Landbase is the stronger choice for revenue teams that need a qualified pipeline, not another infrastructure layer to assemble, maintain, and govern. One platform. One prompt. End-to-end GTM execution. See how Landbase works.

Frequently Asked Questions

How much does Lyzr cost per month?

Lyzr's published subscription tiers are free Community, $19 per month Starter, and $99 per month Pro (or $79 per month billed annually). Enterprise Cloud and On-Premise plans are custom-quoted. Production agent runs are billed separately at $0.08 per run on Lyzr Cloud and $0.03 per run on VPC or on-premise deployments, per Lyzr's pricing page.

Is there a free version of Lyzr?

Yes. The Community plan is free and includes 500 credits per month, 1 builder license, a 100 MB knowledge base, and access to the base model set, per the Lyzr documentation. The Community plan is designed for prototyping and individual developer use rather than production workloads.

What is the difference between Lyzr Cloud and Lyzr VPC pricing?

Lyzr Cloud runs on the vendor's managed infrastructure at $0.08 per agent run. Lyzr VPC runs within the customer's cloud environment at $0.03 per agent run. The VPC and on-premise rates are meaningfully lower per run because the customer brings the compute and infrastructure, while Lyzr Cloud bundles hosting into the run price. For high-volume workloads, VPC typically delivers better unit economics once you factor in the saved run rate against the additional infrastructure you manage.

Are LLM costs included in Lyzr pricing?

No. Lyzr's pricing page specifies that LLM costs are "billed separately at transparent, pass-through usage rates." Token costs to OpenAI, Anthropic, Google, Bedrock, or a self-hosted model accrue directly to the model provider and are not included in Lyzr's credits or agent-run pricing.

How are Lyzr credits used?

Credits meter certain platform operations at build-time and during agent execution, typically around knowledge base operations, tool calls, and specific premium features. Each tier includes a monthly credit bundle (500 on Community, 2,000 on Starter, 10,000 on Pro, unlimited on Enterprise). Additional credits can be purchased one-time from $10 for 1,000 credits up to $5,000 for 500,000 credits, providing flexibility for usage spikes.