Relevance AI

No-code platform for building, testing and running teams of AI agents, guided by its Invent builder agent.

$19 per month (Pro, billed annually)No-code AI agent builder and multi-agent workforce platform

No user reviews yet. Be the first to rate it.

Visit website
Relevance AI website screenshot
Relevance AI website

Quick Facts

DeveloperRelevance AI (OnSearch Pty Ltd)
Websiterelevanceai.com
Agent typeNo-code AI agent builder and multi-agent workforce platform
Best forSales, revenue operations and operations teams that want to build custom agents and multi-agent workflows without engineers.
Not ideal forDevelopers who want a code-first framework, or buyers who need a free plan to experiment.
Pricing modelSubscription with usage-based Actions and Vendor Credits
Starting price$19 per month (Pro, billed annually)
Pricing detailsPro $19 per month billed annually or $29 monthly, with 2,500 Actions a month and 2 build users. Team $234 per month billed annually or $349 monthly, with 7,000 Actions a month, 5 build users and 45 end users. Enterprise is custom. Extra Actions cost $80 per 1,000 and Vendor Credits $20 per 10,000. The Free plan has been retired.
Free planNo
ChannelsWeb app, email, Slack, phone and meeting agents (Enterprise)
What it does on its ownRanges from assisted tasks to self-running agents triggered by events, with optional human-in-the-loop approvals.
IntegrationsSalesforce, HubSpot, Gmail, Google Calendar, Google Sheets, Slack, Zoom, LinkedIn, Perplexity, MCP
PlatformsWeb
Compliance and securitySOC 2 Type II, GDPR; SSO/SAML, role-based access, audit logs, data residency options
Details last verified28 October 2026
CategoriesMulti-Agent Platforms, No-Code Agent Builders, Sales & Lead Qualification, Workflow Automation
User ratingNo reviews yet

Relevance AI’s Pro plan costs $19 a month on annual billing and includes 2,500 agent Actions. The Sydney-founded company sells a no-code platform for building what it calls an AI workforce: teams of specialist agents that handle sales research, lead qualification, support triage and back-office tasks.

Its newest piece, Invent 2.0, is itself an agent that designs and builds those teams. Customers named on its site include Canva, Autodesk, Confluent and Databricks.

How It Works

A user describes a job or shares a standard operating procedure. Invent asks about the judgment calls, connects the systems involved, such as Salesforce, HubSpot, Gmail or Slack, and sets up a team of agents with the tools each needs. Before agents touch real work, they must pass quality checks. Once live, Invent samples runs, flags drift and suggests changes to prompts, tools or models.

Agents can be triggered by events such as a new form submission and chained into multi-agent workflows. Relevance AI offers access to many LLMs through a router that tries to use the cheapest model that passes evaluations. Every run is traced with its cost, and tool access can be governed through an MCP gateway. The company frames autonomy in levels, from assisted chat up to self-improving agents.

Pricing

Pro costs $19 a month billed annually or $29 monthly, with 2,500 Actions a month and two builders. Team costs $234 a month annually or $349 monthly, with 7,000 Actions, five builders and 45 end users. Enterprise is custom and adds unlimited agents, calling and meeting agents and a dedicated account manager. Extra Actions cost $80 per 1,000. Model usage is billed separately as Vendor Credits at $20 per 10,000. The free plan has been retired.

What to Watch Out For

Costs run on two meters. Actions count agent steps and Vendor Credits cover model and data costs, so a busy agent can burn through both faster than expected. Without a free plan, testing means paying for Pro. Some features buyers may assume are standard, including phone agents, sit on Enterprise. Results quoted on the homepage, such as one customer’s $7 million in pipeline, come from hand-picked case studies.

Recent News

On September 29, 2026, the company launched Invent 2.0, which builds and manages agent workforces after a months-long design partner program. In September 2025 it split its credit system into Actions and Vendor Credits, a change it said would raise prices, and later retired its free plan.

Features

  • Invent agent that interviews you and builds a team of specialist agents
  • Pre-built agents for sales, customer success, marketing, HR, support and research
  • Event-based triggers and multi-agent workflows
  • Evaluations that sample live runs against quality checks
  • LLM router that picks the cheapest model that passes evals
  • Full run tracing and per-task cost visibility
  • MCP gateway for governed tool access

Pros and Cons

Pros

  • Builds agents from a plain description of the process
  • Works with many LLMs rather than one vendor
  • Built-in testing, monitoring and cost tracking
  • Low-cost entry plan with published prices
  • Enterprise customers include Canva, Autodesk and Confluent

Cons

  • Free plan retired, so testing requires a paid plan
  • Two usage meters (Actions and Vendor Credits) make costs harder to predict
  • Calling and meeting agents and unlimited agents sit on the Enterprise plan
  • Result figures on its site come from selected customers

User Reviews

No reviews yet. Be the first to rate Relevance AI.

Write a Review