Relevance AI: assembling a team of AI agents without code
Relevance AI is a low-code and no-code platform for building AI agents and agent teams that complete tasks on their own, from lead qualification to customer support. It describes itself as the home of your “AI Workforce”.
Data as of October 22, 2026
Key figures
| Developer | Relevance AI (OnSearch Pty Ltd) |
|---|---|
| Type | Low-code |
| Open source | No |
| License | proprietary |
| Suggested level | Intermediate |
| Use cases | multi-agent, automation, workflows |
| Self-hostable | No |
| Business model | Paid |
| Cloud offering | Yes |
| Multi-agent | Yes |
| MCP support | Yes |
| Bring your own model | Yes |
| Official site | Official site · Documentation |
Agents, tools and Workforces
Relevance AI, operated by OnSearch Pty Ltd, is built on three pieces:
- agents: an AI agent gets a role, instructions, a knowledge base and tools, then acts on its own or when asked;
- tools: no-code sequences of steps (for example web search, API calls, text processing, links to your apps) that agents can use;
- Workforces: a visual canvas for organizing several specialized agents into a multi-agent system that works together on a task.
The platform works with models from OpenAI, Anthropic, Google (Gemini), Azure or through OpenRouter. It provides an MCP server so you can drive your agents from a compatible assistant, and an MCP client to give your agents external tools.
What you can actually automate
The documentation cites customer support (handling routine requests, escalating complex cases to a person), sales prospecting (prospect research, lead qualification, first outreach), drafting routine content, data analysis and calendar management. For sensitive actions, such as an email sent to a customer, add a human-in-the-loop review step; the Can I trust an agent with this? tool helps you weigh the risk.
Pricing as of October 22, 2026
Relevance AI bills in Actions (one per tool run) and Vendor Credits (model costs at wholesale prices). Using your own API keys avoids spending those credits. The free plan is gone and the public pricing page only shows Enterprise, quoted on request; the documentation, however, lists Pro ($19 per month billed annually, $29 monthly, 2,500 Actions) and Team ($234 per month billed annually, $349 monthly, 7,000 Actions). Check which plans are actually offered when you sign up.
Who it fits, and the alternatives
- Freelancer: with no free plan, even a trial means paying for a subscription.
- Small business: a good match for sales or support teams that want agents ready to work.
- Technical team: no code access and no self-hosting.
Strengths: quick to learn, visual multi-agent orchestration, MCP in both directions. Limits: two-meter billing that is harder to predict, and a closed platform. Compare it with Zapier Agents or Make to stay no-code, and with n8n to keep control over hosting.
Frameworks of the same type
- Microsoft Copilot Studio · chatbot, workflows, automation
- Dify · chatbot, RAG, workflows
- Langflow · RAG, multi-agent, workflows
- n8n · automation, workflows, RAG
- Gemini Enterprise Agent Platform (ex-Vertex AI Agent Builder) · multi-agent, RAG, workflows
- Botpress · chatbot, RAG, automation
Frequently asked questions
Does Relevance AI have a free plan?
Not anymore as of October 22, 2026: the Free plan has been retired for new sign-ups and existing accounts alike. You need a paid plan or the Enterprise offer.
How much does Relevance AI cost?
The public pricing page only highlights Enterprise, quoted on request. As of October 22, 2026, the official documentation lists Pro at $19 per month billed annually or $29 monthly (2,500 Actions and $20 in Vendor Credits per month) and Team at $234 per month billed annually or $349 monthly (7,000 Actions and $70 in Vendor Credits).
What are Actions and Vendor Credits?
An Action is counted each time an agent runs a tool. Vendor Credits cover the cost of models and third-party services, passed through at wholesale prices with no markup. If you bring your own API keys, those calls do not use Vendor Credits.
Can I self-host Relevance AI?
No. It is proprietary software available only as an online service. If self-hosting matters to you, n8n, Dify and Langflow all offer it.
Data sources
Data as of October 22, 2026
Report an error
Does a figure, a date or a fact look wrong? Tell us: every report is checked.