Relevance AI
トレンドRelevance AI is a no code AI workforce platform where companies build, train, and manage teams of AI agents that take on repetitive sales, research, support, and operations work under human supervision.
概要
Relevance AI is one of the most complete platforms for building a supervised AI workforce in 2026: instead of one chatbot, companies assemble teams of specialized agents that hand tasks to each other, draw on company knowledge, and act inside everyday tools. Best for operations, sales, and support leaders who want recurring work automated under human supervision with full audit trails. Each agent is built without code by giving it instructions, knowledge sources, and tools such as CRM lookups, web research, document analysis, or message drafting, and prebuilt templates for inbound lead qualification, prospect research, support triage, and report preparation deliver value in the first day. The platform connects to HubSpot, Salesforce, Slack, Google Workspace, and popular databases, and agents run on event triggers or schedules. Activity logs for every agent plus analytics on completed work and human interventions let teams expand automation with confidence. A free plan with 1,000 credits and 200 actions per month supports evaluation, while paid plans start around $29 per month and scale to enterprise deployments with security reviews and dedicated support. The honest limits: complex agent teams need careful setup time, and higher volume usage gets expensive. A 4.4/5 rating with 4/5 marks for ease of use, value for money, and support reflects a platform that trades some polish for breadth, and GDPR plus SOC 2 compliance with region selection covers regulated teams.
主な機能
メリット
- +Complete platform for multi agent operations
- +Detailed audit logs build trust in automation
- +Templates deliver value in the first day
デメリット
- -Complex agent teams need careful setup time
- -Higher volume usage gets expensive
おすすめ用途
連携と互換性
Frequently Asked Questions
What is Relevance AI used for?
It is used to build AI agent teams that take on recurring work such as lead qualification, prospect research, support triage, and reporting, acting inside company tools under human supervision.
Do I need to code to build agents?
No. Agents are configured with instructions, knowledge sources, and tools through a no code interface, though developers can extend them with custom tools and APIs.
How much does Relevance AI cost?
A free plan with 1,000 credits and 200 actions per month supports evaluation. Paid plans start around $30 per month and scale to enterprise tiers with security reviews and dedicated support.
How is it different from a chatbot?
Chatbots answer questions. Relevance AI agents complete multi step work: they research, update CRMs, draft documents, and hand tasks between each other, with activity logs for auditing.
Can agents use our company knowledge?
Yes, agents can be grounded in your documents, wikis, and databases, and they can connect to tools like HubSpot, Salesforce, and Slack to act on that knowledge.
Can Relevance AI agents run on a schedule?
Yes. Agents can be triggered by events or run on schedules, and every run is recorded in an activity log. Teams typically start with scheduled prospect research or support triage, then expand to event triggered agents once the workflow proves itself.
What integrations does Relevance AI support?
Relevance AI connects to HubSpot, Salesforce, Slack, Google Workspace, and popular databases, with event triggers and scheduled runs for activation. Activity logs capture each action across connected tools, so managers can audit what the automation did and where humans stepped in.
What is a credit in Relevance AI and how far does it go?
Credits are the usage currency: each agent run, tool call, or LLM step consumes credits based on the work performed. The free plan includes 1,000 credits and 200 actions per month, which is enough to prototype one or two agents. Paid plans from $29 per month raise the ceiling as your agents move into daily operation.
How long does it take to build a first agent?
Most users assemble a working agent in under an hour using templates and the no code tool builder: pick a role, attach knowledge, add tools such as search or data lookup, and test in the preview console. The first pass is fast, and iteration over prompts and tools is where answer quality actually comes from.
Does Relevance AI replace tools like Zapier or n8n?
Not exactly. Zapier and n8n excel at linear trigger and action automations between apps, while Relevance AI builds agents that reason, use tools, and handle open ended tasks. Many teams run both: classic automation moves data between systems, and Relevance AI agents sit on top for the steps that need judgment.
What happens when I hit the free plan limit?
Agents pause until credits or actions reset on the monthly cycle, or you upgrade to a paid plan from $29 per month to continue. Prototypes usually stay comfortably inside the free allowance, so the limit tends to bite exactly when an agent proves useful, which is the natural upgrade moment.
Can multiple agents work together?
Yes. You can chain agents so the output of one becomes the input of another, or run a lead agent that delegates subtasks to specialist agents with different tools and knowledge. This multi agent pattern suits workflows such as research, content pipelines, and lead qualification where a single agent would be too broad.
評価の内訳
対応言語
データプライバシーとセキュリティ
GDPR and SOC 2 compliant with region selection