Relevance AI

What it does

Done-for-you service

PartlyYou run the software, but a Relevance deployment team builds the first agents with you and Enterprise includes custom implementationAn embedded deployment team gets you live, then trains your team to build without us. All on top of Relevance's enterprise-grade platform.relevanceai.com ↗

How much it does unattended

Runs autonomously

YesAgents are positioned to complete tasks without a human in the loop; approval modes are configurable per connectionRelevance AI is a low/no-code platform where you can build AI agents and multi-agent teams that autonomously complete tasks, much like human…relevanceai.com/docs/get-started/introd… ↗

Multi-agent

YesWorkforces link multiple agents on a visual canvas with agent-to-agent handoffs; 1 Workforce on Free, unlimited on paidMulti-agent teams where specialized AI agents collaborate on complex tasksrelevanceai.com/docs/get-started/introd… ↗

Agent permissions

YesPer-connection approval modes: Auto Run, Approval Required, or Let Agent Decide, with a Max auto runs capThis mode forces the agent to seek explicit approval before executing an action.relevanceai.com/docs/build/workforces/w… ↗

Which models it runs on

Claude

YesAnthropic models served via Anthropic's API with no training on your data; Claude thinking/effort controls exposed in agent settingsAnthropic models are offered through their API, and once again, no data is stored or trained on during this process.relevanceai.com/docs/admin/security ↗

GPT

YesOpenAI GPT models are available by default via vendor credits, or with your own OpenAI keyThe below table describes the LLM models available in Relevance AI and their respective vendors when no API key is provided.relevanceai.com/docs/admin/security ↗

Open models

YesOpen-weight models reachable through the OpenRouter integration; the vendor also hosts LLaMA-family models in its own cloudOpenRouter models are a unified access point to a wide variety of open models like Meta’s LLaMA family, Google Gemini models, and others.relevanceai.com/docs/integrations/llm-i… ↗

Model choice

YesYou pick the model per agent, on every tier, plus a fallback model from another providerUse any LLMs with creditsrelevanceai.com/docs/get-started/pricing ↗

Where you use it

In your editor

PartlyNo native IDE extension; its MCP server connects from Cursor, VS Code/Copilot, Windsurf and Zed to build and run agentsThe Relevance AI MCP server gives any MCP-compatible AI client direct access to your agents, tools, and knowledge.relevanceai.com/docs/integrations/mcp/m… ↗

On the command line

PartlyNo first-party CLI; a Claude Code plugin and Codex MCP registration drive it from a terminalThe Relevance AI Claude Code plugin is the fastest way to build Relevance AI agents from your terminalrelevanceai.com/docs/integrations/mcp/c… ↗

In a browser

YesHosted builder at app.relevanceai.com plus a chat interface on desktop, mobile and browserUse AI Agents and Workforces in a chat interface on desktop, mobile, and browserrelevanceai.com/docs/get-started/chat/i… ↗

Whose machine it runs on

Self-hosted

NoMulti-tenant SaaS with logical separation; single-tenant is stated as not yet availableRelevance operates on a multi-tenanted architecture where customer data is logically separatedrelevanceai.com/docs/admin/security ↗

Not established: Open source

What it costs to run

How it meters

YesTwo meters on top of a seat-limited plan: Actions (one tool run each) and Vendor Credits (LLM/tool cost at wholesale)An Action is a single run of a Tool. Each time a tool runs, it counts as an action — whether it's a simple task like sending one email or running a…relevanceai.com/docs/admin/subscription… ↗

Free tier

Yes$0 Free plan with 200 Actions/month and 1,000 one-time Vendor Credits, no card requiredNo, Relevance AI's free plan doesn't require a credit card.relevanceai.com/docs/get-started/pricing ↗

API access

YesRelevance API key plus Python/JavaScript SDK to trigger agents and run tools; API triggers documented on all tiersUse the Python or JavaScript SDK to trigger agents, run tools, or manage resources programmaticallyrelevanceai.com/docs/get-started/core-c… ↗

MCP server

YesHosted MCP server at mcp.relevanceai.com with OAuth and a server-side Run-Only mode; it can also consume external MCP serversConnect to Relevance AI from Claude Desktop, Cursor, VS Code, ChatGPT, and other MCP-compatible AI clients.relevanceai.com/docs/integrations/mcp/m… ↗

Bring your own key

PartlyBring your own LLM API keys to bypass Vendor Credits, but only on paid plansYes! You can bring your own API keys to bypass Vendor Credits entirely. This is only available on our paid plans and is not available to free users.relevanceai.com/docs/admin/subscription… ↗

Buying it for a team

A company can buy it

YesSubscriptions are bought at Organization level; Enterprise is a sales-led custom planSubscriptions are applied at the Organization level, not to individual users or projects.relevanceai.com/docs/get-started/pricing ↗

Seat model

YesHard seat ceilings per plan: 1 Build User on Free, 2 on Pro, 5 Build plus 45 End Users on Team, unlimited on Enterprise5 Build Users · 45 End Usersrelevanceai.com/docs/get-started/pricing ↗

Pooled budget

YesOne organisation-wide balance of Actions and Vendor Credits shared by all membersEveryone in the Organization shares the plan's Actions, Vendor Credits, and feature access.relevanceai.com/docs/get-started/pricing ↗

Admin controls

PartlyApproval gates, spend and usage limits on all plans; RBAC and fine-grained asset controls are Enterprise-onlyRole-based access control (RBAC) for organizations and projects is available for Enterprise customersrelevanceai.com/docs/admin/security ↗

Audit log

PartlyAudit logs are an Enterprise feature, streamed as OTEL records to your own S3 bucket; marked ✗ on Free, Pro and TeamAudit logs provide a complete record of activity across your organization for security monitoring, compliance reporting, and operational visibility.relevanceai.com/docs/enterprise/streami… ↗

Single sign-on

PartlySAML 2.0 and OIDC via WorkOS, Enterprise plans only, set up with a Relevance representativeSSO is available for Enterprise customers only.relevanceai.com/docs/enterprise/sso-set… ↗

What happens to your code

Opt out of training

YesNo training on customer data by default, and the default-model vendors are contracted not to train eitherWe do not use your data to train our models or improve our services unless you have a specific partnership agreement with usrelevanceai.com/docs/admin/security ↗

Data residency

YesChoose US (N. Virginia), EU (London) or AU (Sydney) at signup; the region cannot be changed afterwardsData is stored in Australia, the US, or the EU/UK, based on your selection at signup.relevanceai.com/docs/admin/security ↗

Getting out

YesExport in CSV, Excel or JSON at any time; Vendor Credits roll over indefinitely while subscribed, but plan Actions reset each renewalYou can export your data at any time in standard formats (CSV, Excel, JSON)relevanceai.com/docs/admin/security ↗

Certifications

YesSOC 2 Type II and GDPR on every tier; reports and questionnaires only for Enterprise under NDARelevance AI is SOC2 Type II compliant, and we operate in a multi-region environment.relevanceai.com/docs/admin/security ↗