Flowise v2: visual agents with parallel tool use and configurable memory types
In one sentence Flowise v2 introduces sequential and parallel tool use in agents, multiple memory types (buffer, summary, vector), visually configurable agent loops, and LlamaIndex support.
Flowise is an open source visual builder for building LLM applications and agents via drag-and-drop, similar to n8n but specialized for AI flows. With version 2, agents become much more capable: they can use tools in parallel, have different memory types, and agentic loops are configurable without writing code.
You can build an agent with short-term memory (buffer), summary memory, and semantic search (vector) by selecting them from the canvas. Results are visible in real time in the integrated test panel.
Flowise distinguishes itself from Dify through closer alignment with the LangChain and LlamaIndex ecosystem: those already familiar with these frameworks find the concepts familiar in the visual canvas.
Companies
FlowiseAI
Tools
Flowise, LangChain, LlamaIndex
Tags
Sources