Skip to main content

Rate limits

Written by Product Management

This article is auto-synced from its in-app version in Tai.

Company admins only. This page is part of Settings, visible to users with the Company Admin role.

Rate limits cap how many requests per minute your company, its individual users and API keys, and its REST-connected agents may make. Settings → Rate limits is where you see the ceiling Nezasa has granted you, tighten it if you want to, and set the shared default every user and API key draws on.

The Rate limits page under Settings, showing the allocated limit, company limit and per-caller default fields above the per-agent limits section, which lists no REST-connected agents yet

The three numbers

  • Allocated limit — the ceiling Nezasa has granted your company. You can see it here, but only Nezasa can raise it; contact your Nezasa representative if you need more headroom.

  • Company limit — an optional lower cap you set for yourself, shared by everyone in your company. Leave it empty to use your full allocated limit. You cannot raise it above your allocated limit. If Nezasa later lowers your allocation below a company limit you had already set, your value is kept — but the allocated limit is what applies, and the page shows a note saying so.

  • Per-caller default — how many requests per minute any single user or API key may make. This is not limited to one feature: the same allowance covers chat messages, report runs, knowledge queries and REST API agent calls, so lowering it also lowers how many chat messages a person can send in a minute.

Per-agent limits

An agent connected over the REST API can have its own limit instead of the per-caller default — useful when one integration needs more throughput than everyone else. Every REST-connected agent is listed in the table on this page. Type a number to give one its own limit, change it whenever you like, and clear the field to put the agent back on the per-caller default.

An agent's own limit is shared by everyone who calls that agent, not granted per caller — size it for the integration's total traffic, not per key. Every agent's calls, whatever limit applies to them, still count against your company limit above.

When you'd use this

Lower the company limit if you want your own tenant-wide ceiling below what Nezasa has allocated — for example, to leave headroom for other Nezasa products sharing the same allocation. Lower the per-caller default if one busy integration or user should not be able to crowd out everyone else's chat and report usage. Give a REST-connected agent its own higher limit when a single integration needs more throughput than the per-caller default allows.

Who can change what

Only an admin of your own company can change these limits. Nezasa staff administering your company can see them — they need to, to explain a throttle — but cannot change them on your behalf, and the page says so while they are administering.

Tips and limits

  • The per-caller default is not per feature — it is the one allowance chat, reports, knowledge and REST API calls all draw from for that caller.

  • A per-caller default set above your company limit is accepted and stored, but the company limit still applies in practice — the page shows a note when that's the case.

  • Only agents with a REST API channel appear in the table. If an agent you want to limit isn't listed, enable that channel under the agent's Configuration → Channels and add a contract version — see Setting up an agent's REST API.

  • Built-in agents (the ones Nezasa ships) never appear in the table — you cannot change their channels or limits.

  • Clearing a field never removes the cap. Clearing the company limit puts you back on your full allocated limit; clearing the per-caller default puts that caller allowance back to Nezasa's standard value.

Related

Did this answer your question?