AgentWorks vs Relevance AI: EU AI Workforce Compared

Both AgentWorks and Relevance AI let you build a team of AI agents that research, draft, and act across your tools. The practical difference comes down to where your data lives, how you pay for model usage, and how much governance is built in from day one.

Two ways to build an AI workforce

Relevance AI is a well-known platform for assembling AI agents and multi-agent teams, with a large template library and a strong focus on sales and go-to-market automation. AgentWorks takes the same core idea, a coordinated AI workforce, and builds it for European organisations first.

That framing shapes every decision. AgentWorks is built in the Netherlands and founded in 2025, with EU data residency, at-cost token billing, and EU AI Act-ready governance treated as defaults rather than add-ons. If you are a European team weighing which platform to standardise on, those defaults are usually the deciding factor.

Both platforms give you pre-built AI agents to start fast. AgentWorks ships 50+ ready-to-use agents from the Free plan, so you can test real work before committing budget.

EU data residency and governance

The clearest difference is where your data is processed. AgentWorks offers EU data residency and uses EU model endpoints where they are available, so prompts and documents stay within European infrastructure by design. For teams under GDPR, that removes a recurring procurement headache.

Governance is built into the runtime, not bolted on. Every agent carries a per-agent risk classification, state-changing actions require human-in-the-loop approval, and every step writes to an immutable, append-only audit trail you can export as CSV or JSON. Personally identifiable information is masked at the gateway before any model sees it, and AgentWorks runs on no-training, zero-retention model contracts so your content is never used to train third-party models.

On the regulatory side, AgentWorks is EU AI Act-ready rather than blanket "compliant" — actual risk always depends on your specific use case, and the platform gives you the per-agent risk classification and human oversight controls the Act expects. A DPA is available on request. You can read the full picture on the compliance page.

Transparent at-cost token billing

Pricing models diverge sharply here. AgentWorks bills tokens at cost plus 10% from a single transparent € wallet. There are no credit conversions to decode: you see live per-run spend, and you can set budgets at org, team, and user level so nothing runs away from you.

The AUTO router is what makes this efficient. Instead of paying a flat premium for one frontier model, AUTO sends each message to the cheapest model capable of handling it. Simple tasks route to fast, low-cost models; hard reasoning routes to the strongest ones. You get quality where it matters without overpaying for routine work.

The plan structure is straightforward. Free costs €0 and includes a €5 one-time credit, 50+ agents, up to 3 integrations, a personal knowledge base, and the AUTO router. Pro is €39/month with a €10/month balance included, plus custom agents, the visual workflow builder, scheduled agents, and organisation knowledge. Team is €49/seat/month with shared chat, knowledge, and admin. Enterprise is custom, adding engineer-built agents, self-hosting or private cloud, SSO/SAML, an SLA, and local models. Full details are on the pricing page.

Models, chat, and document output

AgentWorks is genuinely multi-model. You can switch between GPT-5 and GPT-5 mini, Claude Opus, Sonnet and Haiku, Gemini Pro and Flash with up to 1M-token context, and Mistral Large — mid-conversation, without losing context. Image generation runs on Gemini image (Nano Banana) and OpenAI image, and Enterprise customers can run local or small language models on their own infrastructure.

The chat experience is built for real output. Alongside the models you get web search, image generation, cited Deep Research, code execution, and grounding in your company knowledge. You can create and export Word, PowerPoint, Excel, and PDF files in a live canvas, then open them directly in Google Drive or OneDrive. That closes the gap between "the agent answered" and "I have the deliverable."

Multi-agent pipelines and knowledge

Where a single chat ends, multi-agent pipelines begin. You can chain agents into flows like research → draft → review → publish, run them on a daily, weekly, or monthly schedule on Pro and above, or trigger them from a webhook. Every step is logged and carries its own risk class, so an automated pipeline stays as auditable as a manual one.

Grounding those agents is a proper knowledge and RAG layer. Upload PDF, DOCX, TXT, or CSV files, or connect URLs, Notion, and Confluence. Content is embedded with pgvector, answers come back with citations, and the system says "I don't know" when the answer isn't in your knowledge base rather than inventing one — a meaningful safeguard for regulated work.

Both platforms connect widely. AgentWorks integrates with Slack, Microsoft Teams, Gmail, Google Workspace, Google Drive, OneDrive, SharePoint, Salesforce, HubSpot, Pipedrive, Notion, Confluence, Jira, Asana, Monday, Calendly, GitHub, GitLab, and Exact Online, plus MCP servers and a REST API with inbound webhook triggers.

Summary: AgentWorks and Relevance AI both build AI agent teams, but AgentWorks is engineered for European organisations: EU data residency and EU model endpoints, at-cost token billing from one transparent € wallet with an AUTO router that picks the cheapest capable model, and EU AI Act-ready governance with per-agent risk classes, human approval, and an immutable audit trail — all available to try free with 50+ agents.

Frequently asked questions

Is AgentWorks a direct alternative to Relevance AI?

Yes. Both let you build and run teams of AI agents across your existing tools. AgentWorks differentiates on EU data residency, transparent at-cost billing with AUTO model routing, and governance features built for the EU AI Act, which makes it a natural fit for European teams evaluating Relevance AI.

How does AgentWorks pricing compare?

AgentWorks charges tokens at cost plus 10% from a single € wallet, with live per-run spend and budgets at org, team, and user level. Plans start at Free (€0 with a €5 credit and 50+ agents), then Pro at €39/month and Team at €49/seat/month, both including monthly balance. Enterprise is custom.

Can I keep my data in the EU with AgentWorks?

Yes. AgentWorks offers EU data residency and uses EU model endpoints where available, masks PII at the gateway before any model call, and runs on no-training, zero-retention contracts. A DPA is available on request. See the compliance page for the full detail.