Docs

Documentation

Build a workflow

Create a complete AI workflow in minutes: select providers, configure fallbacks, attach structured outputs, and define cache fields.

Create a new workflow

  1. Open Workflows in your project sidebar and click Create Workflow.
  2. Enter a name. Choose a consistent convention such as team-purpose-environment (e.g. support-summarizer-prod).
  3. Select the primary provider and model from the dropdowns.
  4. Optionally configure fallback providers: secondary and tertiary options that activate when the primary fails or times out.
  5. Attach a structured output schema if you need strict JSON responses.
  6. Add cache fields (comma or newline separated) to surface business identifiers in API responses.

Use the connected nodes icon in the sidebar to open Workflows; the form repeats the provider (server glyph) and model (chip icon) selectors so you recognise them from other views.

Configure fallback behaviour

ModelRiver automatically retries with the backups you configure when the primary provider fails or times out.

  • Every attempt appears in meta.attempts, including reasons for failures when available.
  • Use backups to mix vendors (for example, primary OpenAI, fallback Anthropic) or to tier model sizes.
  • Set timeout thresholds per provider to control when fallback activates.

Workflow naming conventions

ConventionExampleBest for
team-purpose-envsupport-summarizer-prodMulti-team organisations
feature-variantonboarding-v2A/B testing workflows
product-tierenterprise-analysisTiered service levels

Test a workflow

  1. Open the Playground from the project sidebar (Projects → Playground).
  2. Pick a workflow from the dropdown.
  3. Provide a JSON payload containing the fields your workflow expects: commonly a messages array and optional metadata.
  4. Click Send Request. ModelRiver executes the workflow without persisting a log entry, and the response appears inline.
  5. Iterate on prompts, cache fields, or structured outputs based on the preview without impacting production analytics.

The play button icon in the sidebar marks the Playground entry point; inside you'll see the same connected nodes icon on the workflow selector so you know you're running an existing configuration.

Next steps