Relevance AI
UnknownEnterprise platform for building, running, and managing AI agents at scale with orchestration, evaluations, integrations, and production controls.
Информация об инструменте
Что такое Relevance AI?
Relevance AI is an enterprise-grade platform that enables teams to build, run, and manage specialist AI agents at scale. It provides the runtime, models, integrations, knowledge, multi-agent orchestration, evaluations, and governance controls needed to put agents into production and keep them there.
Основные функции
Multi-agent orchestration
Node-and-edge graphs with agent, tool, and condition nodes; handoffs, sub-agent calls, parallel and branching runs, and a visual drag-and-drop builder.
Evals and quality oversight
Pre-publish eval sets and thresholds, production sampling and alerts, email and Slack escalations, condition-to-action alerts, PII data masking, and version management (draft, active, pinned).
Knowledge and memory
Long-term memory (project and user), compaction and observational memory, hybrid RAG with reranking, automatic syncing with key data sources, and custom metadata filled by agents; shared context layer for tone of voice, business context, and knowledge.
Tools and actions
Tested integration steps, sub-agents and MCP servers, sandboxed Python and JavaScript, per-tool approval rules, and retry and execution limits.
Triggers and scheduling
Recurring schedules, webhooks, hundreds of pre-built events, custom integration cron triggers, bulk runs across datasets, and call via API.
Models and LLM routing
Support for any major provider plus bring-your-own, fallback model on failure, temperature and max output tokens, reasoning and thinking modes, strict and parallel tool calling, and model switching after N tokens; run on the cheapest model that passes evals.
Channels and sharing
Voice agents, embeddable widgets and chat bubble, public and marketplace listing, shareable links, welcome and starter messages, and suggested replies.
Workflow builder
n8n-style visual workflow builder and drag-and-drop canvas to compose agents, tools, and approvals; Build with AI (Invent) that generates prompts, tools, and evals from plain-language descriptions.
Enterprise security and controls
Security and data privacy features, access and controls, monitoring and oversight, autonomy step limits, approval rules, cost-based pauses, and governed tool access via MCP gateway.
Integrations
Connect to 1,000+ apps with MCP gateway, tested integration steps, and triggers such as inbound email, Salesforce leads, Slack messages, webhooks, and CRM updates.
Варианты использования
- #1
Sales agents
Specialist agents for research and enrichment, pre-meeting prep, post-call actions, meeting scheduling, outbound prospecting, forecast roll-up, deal review, and proposal building.
- #2
Customer success and support
Agents for escalating at-risk renewals, triaging support tickets, resolving tickets, and providing 24/7 coverage and automated conversations.
- #3
Marketing and operations
Agents for nurturing qualified inbounds, personalized multi-channel engagement, invoice matching, chasing overdue invoices, and contract review.
- #4
HR and research
Purpose-built agents for human resources workflows and research and enrichment from live dynamic sources.
Тарифные планы
No public pricing details on official pages; Talk to sales and Book a demo options are offered.
Альтернативы
Часто задаваемые вопросы
What is Relevance AI?
What can I build with multi-agent orchestration?
Does Relevance AI support evaluations and quality controls?
Which platforms and channels are supported?
How is pricing structured?
What integrations are available?
Relevance AI Контакты и социальные сети
- Поддержка
- Веб-сайт:
- Twitter
- YouTube
- LinkedIn
- GitHub
Отзывы
Поделитесь своим опытом использования этого инструмента...