Public LLM

Not every agent needs a private model

For agents that don't handle regulated or highly sensitive data, a hosted model provider gets you live faster with no infrastructure to manage. Lira supports public model providers alongside private deployment, chosen per agent.

No credit card required · 100 free tokens · Cancel anytime

When should you use a public LLM instead of a private one?

A public, hosted model provider is the right default when speed to deployment matters more than data residency — most sales and support use cases fall here. Private deployment becomes necessary specifically when regulation, contractual terms, or internal policy prohibit sending conversation data to a third-party API. Lira lets you make this choice per agent rather than committing the whole platform to one or the other.

Capabilities

What you get

Multiple provider support

Connect the hosted model provider your team already has an account with.

Fast to configure

No infrastructure to provision — an agent can be live within minutes.

Per-agent model choice

Run some agents on a public provider and others on a private LLM in the same account.

Fallback provider

Configure a second provider as automatic fallback if the primary is unreachable.

Same feature set

Knowledge base, workflows and tool calls work identically regardless of model choice.

Usage visibility

Track usage and cost per provider from the same billing view.

Use cases

Who this is built for

Fast pilot deployments

Get a first agent live quickly to prove value before deciding on a private deployment.

Non-regulated use cases

Use a hosted provider for sales and support flows with no data-residency requirement.

Resilience via fallback

Keep agents running even if a primary provider has an outage.

FAQ

Common questions

Can we switch from public to private later?

Yes — model choice is an agent-level setting, so moving to a private LLM later doesn't require rebuilding the agent.

Which providers are supported?

Configuration is scoped per your account — talk to our team about the specific provider you want to connect.

Ready to deploy public llm?

Every new account starts with 100 free tokens — enough to build and test-call your first agent before spending a rupee.

Start free trial →