name: flowise-prediction
description: |
Invoke a Flowise chatflow or agentflow to generate a model response. The
Prediction endpoint is the primary runtime surface for any Flowise-built AI
application. Supports streaming, multi-turn history, runtime overrides, RAG
source documents, tool use, and file uploads.
provider: flowise
api: flowise
authentication:
type: bearer
description: HTTP Bearer using a Flowise API key. The chatflow may also have
its own per-flow API key requirement.
baseUrl: http://localhost:3000/api/v1
operations:
- id: create-prediction
method: POST
path: /prediction/{id}
summary: Send a message to a flow and receive an AI response
notes: |
Body fields include `question`, `streaming`, `overrideConfig`,
`history`, `uploads`, and `form` (for Agentflow V2). When `streaming`
is true the response is delivered as Server-Sent Events.
tags:
- Prediction
- Inference
- Runtime
- Streaming