Visual AI pipeline builder

Build AI workflows without writing glue code

Drag-and-drop 10 AI nodes onto a canvas. Connect them. Run. Watch your pipeline execute in real time with per-node status, outputs, and cost tracking.

Contract Text
LLM Chain
Extract Schema
JSON Output
Execution Output
Completed

{ "recommendation": "REVIEW", "key_risks": ["auto-renewal", "liability cap"] }

Legal Document Analyzer — live demo, no API key needed

How it works

01

Design your pipeline

Drag nodes from the palette onto the canvas. Connect outputs to inputs to define the execution order.

02

Configure each step

Click a node to open its config panel. Write prompts, set schemas, choose models — all inline.

03

Run and observe

Hit Run. Watch nodes animate in real time. Inspect outputs, token usage, and cost per node.

Everything you need

From document ingestion to structured output — in one canvas.

Real-time execution

Watch each node animate from idle → running → complete via Server-Sent Events. No polling, no page refresh.

10 composable nodes

LLM Prompt, LLM Chain, Structured Extractor, RAG Retriever, Conditional Branch, Document Upload, and more.

BYOK — your keys, your data

API keys are encrypted at rest with AES-256-GCM. The worker decrypts them per-execution in memory. Never logged.

Document intelligence

Upload PDFs, DOCX, and CSV files. Chunk, embed, and retrieve relevant passages with ChromaDB vector search.

Variable injection

Reference any upstream output with {{nodeId.output}} in your prompts. Chain complex multi-step reasoning pipelines.

Structured extraction

Define a JSON schema and FlowForge will validate and parse LLM output into typed objects every time.

10 node types, one canvas

Every node has a config panel, real-time status ring, and output preview.

Text Input
Document Upload
Chunker
Retriever
LLM Prompt
LLM Chain
Structured Extractor
Conditional Branch
Text Output
JSON Output

Ready to build?

Free to use. Bring your own API keys. No usage limits.

Create your first workflow