Using Downsizing with Flowise

Prerequisites

A running Flowise instance (self-hosted or Flowise Cloud) with permission to edit chatflows, and a Downsizing account that can make requests (pay-per-use or a connected subscription both work). Grab your Inference ID and API key from the Inference API page.

Manual Setup

Flowise's ChatAnthropicnode's inputs (model, temperature, extended thinking, etc.) don't include a base URL override, and its credential (anthropicApi) only accepts an API key — so it can't be pointed at a proxy. The ChatOpenAI node does expose a Base Pathfield under Additional Parameters, which overrides the OpenAI client's base URL — use that node to route through Downsizing instead:

1. In your chatflow, drag in a ChatOpenAI node (under Chat Models).
2. Click Connect Credential → Create New to add an OpenAI Api credential. In the credential form, set:
   • OpenAI Api Key: your Downsizing API key (starts with dws_).
3. On the node itself, expand Additional Parameters and set Base Path to your Downsizing OpenAI-compatible base path (ending in /v1):

http://localhost:3001/YOUR_INFERENCE_ID/openai/v1

Replace YOUR_INFERENCE_ID with your workspace UUID from the dashboard. Set Model Name to whichever Claude or GPT model you want Downsizing to route — Downsizing selects the actual upstream model itself, so any recognizable name works.

Verify

Open the chatflow's test chat panel and send a message. If you get a response back, the node is wired up correctly. Then check your savings dashboard to see the request appear in your usage metrics.