Lyzr AI

Lyzr AI is an enterprise AI agent platform for teams that need to build, deploy, monitor, and govern production agents, with public materials showing both self-serve Agent Studio subscriptions and usage-based agent-run pricing for cloud, VPC, and on-prem deployment options. It is best evaluated as a managed agent infrastructure layer rather than a bare developer framework.

Current public pricing materials give buyers more concrete inputs than a simple quote-only label. Lyzr's pricing page lists Lyzr Cloud at $0.08 per agent run and Lyzr VPC/on-prem at $0.03 per agent run. LLM costs are described as separate pay-per-usage charges, and VPC/on-prem compute is also described as usage-based. The same public pricing narrative says an agent run can include a knowledge base call, tool call, agent call, memory call, Responsible AI guardrails, and an agent security policy. Because the public record mixes agent-run pricing with subscription-style Agent Studio plans, procurement teams should confirm which model applies to their edition, deployment architecture, and expected workload.

The current docs also list monthly Agent Studio plans: Community at $0/month, Starter at $19/month, Pro at $99/month, and Enterprise as custom. Published included usage is 500 credits/month on Community, 2,000 credits/month on Starter, and 10,000 credits/month on Pro. The same docs list up to 10 agents on Community, 15 agents on Starter, and 25 agents on Pro; up to 5, 10, and 15 knowledge bases respectively; 100 MB RAG storage on Community and Starter; 1 GB RAG storage on Pro; and 7-day traceability and observability logs on these self-serve plans. Extra credit top-ups are listed at $10 for 1,000 credits, $50 for 5,000 credits, $100 for 10,000 credits, $1,000 for 100,000 credits, and $5,000 for 500,000 credits. A separate credit documentation page says 1 credit equals $1 for tracking usage, while model inference, token length, semantic indexing, and multi-agent orchestration can consume credits.

For enterprise buyers, public docs describe custom Enterprise pricing with Lyzr Cloud or on-prem deployment, credits and RAG storage as per plan, unlimited agents and knowledge bases, access to all models plus bring-your-own-model options, all blueprints, team collaboration, Agent Entitlement Protocol, Responsible AI and Hallucination Manager, Agent Eval tools, customizable logs, a 48-hour custom integration SLA, and 24/7 technical support. Those details support Lyzr's production and governance positioning, but they do not independently verify formal compliance certifications, audit report availability, regional data residency, uptime SLA language, or the exact current contract terms for regulated deployments.

Feature positioning should therefore stay precise. Publicly supportable language includes Lyzr Agent Studio, no-code or low-code agent building, RAG/knowledge-base workflows, Text2SQL references in the docs FAQ, orchestration, traceability and observability logs, Responsible AI guardrails, Hallucination Manager, Agent Eval, Agent Entitlement Protocol, cloud deployment, and on-prem or VPC deployment options. Specific names that are not clearly substantiated in the supplied record should be avoided unless they are confirmed in current Lyzr documentation or sales materials. Overall, Lyzr AI is strongest for organizations that want packaged agent deployment and governance with measurable usage units, while teams that need open-ended code-level control may still compare it with LangChain, CrewAI, Semantic Kernel, AutoGPT, or Microsoft Copilot Studio.