The API, shared page, workspace and agents all call the same published version. Change it once and every entry point updates.
Every run has a log tree with execution sequence, latency, tokens and credits, and earlier versions can be restored.
The Insights page shows total calls, success rate, credits, tokens and latency.
Start and End are added by the system; add the rest to the canvas as you need them.
The LLM node accepts image, document, audio and video input when the chosen model supports it, and can be given up to 10 tools, MCP tools included, which the model calls as needed. Knowledge Retrieval supports keyword, semantic or mixed search.
Inside loops only
Conditional branches; loops over an array or a set count, where you can also set intermediate variables, continue or break; batches of up to 10 in parallel and 200 per run.
Process data with Python code, parse uploaded files, clean up text and merge results from several branches.
Query data tables, send HTTP requests, or call a connected API or MCP tool directly.
Get it working in the draft, then publish it to the people and systems that need it.
Debug a single node and inspect its output. Before a full test run, you get a list of any run errors.
Each scenario below uses native nodes only.
Upload a batch of contracts or reports, parse each one, pull out the key points and collect them in a single table.
Built for enterprise processes
When changes take effect, who can use a workflow and where it runs are all up to you.
Edits change only the draft and take effect when you publish. You can also cancel a publication.
Workflows are reviewed before they reach the workspace; employees can use them once approved.
Use the public cloud, or deploy privately in your own environment.
Workflow processes data in fixed steps without a conversation and runs via API, a shared page or the workspace. FlowAgent is for conversations that must follow set steps, and it supports handoff to a human agent.
Agents that search knowledge, query data and call tools until the request is resolved.
Perform advanced tasks and solve problems dynamically
Learn moreReal-time voice agents for support, phone lines and speaking practice.
Real-time voice agents that answer calls and talk on your website
Learn moreMulti-step processes orchestrated visually and exposed as APIs.
No-Code AI Automation Platform
Learn moreEquip your AI Agent with extensive business knowledge
Learn moreEnhance and maximize LLM capability with powerful tools
Learn moreBuild AI Agents with intuitive interface
Learn more