If/Else Nodes

If/Else Nodes enable conditional branching in your workflows. They evaluate a condition and route execution to either the True or False path — only one path executes, never both. Dal Nulla offers two variants: standard If/Else (rule-based) and AI If/Else (AI evaluates the condition).

What is an If/Else Node?

Conditional branching lets you create dynamic workflows that adapt based on content. “If the text mentions landscapes, generate a panoramic video. Otherwise, generate a portrait.” The If/Else Node evaluates a condition on its input and sends execution down one of two paths.

Key behavior: Only the True OR False path executes. The other path is marked as “skipped” and consumes no credits.

Inputs & Outputs

PortDirectionTypeDescription
inputInText/AnyValue to evaluate — from Text, JSON Parser, Fetch, List Selector, Concatenator, or any text-producing node
trueOutAnyTriggers when condition is TRUE → connects to any node type
falseOutAnyTriggers when condition is FALSE → connects to any node type

Two Variants

Standard If/Else

Rule-based evaluation using comparison operators. Best for exact, predictable conditions.

Inspector Controls: | Control | Description | |---------|-------------| | Condition Field | The value to test (from connected input) | | Operator | Comparison type (see table below) | | Comparison Value | The value to compare against |

Operators: | Operator | Description | Example | |----------|-------------|---------| | Equals | Exact match | Input equals “landscape” | | Contains | Substring match | Input contains “error” | | Greater Than | Numeric comparison | Input > 100 | | Less Than | Numeric comparison | Input < 50 | | Is Empty | Check for empty input | Input is "" | | Is Not Empty | Check for any content | Input has text |

AI If/Else

AI-powered evaluation using a natural language condition prompt. Best for subjective or complex conditions.

Inspector Controls: | Control | Description | |---------|-------------| | Condition Prompt | Natural language question for the AI (e.g., “Is this content suitable for children?”) | | Model | AI model to use for evaluation |

The AI reads the input content and the condition prompt, then decides True or False.

How to Use

  1. Add an If/Else Node (or AI If/Else) to the canvas
  2. Connect a text source to the input port
  3. Standard: Set the operator and comparison value
  4. AI: Write a condition prompt describing what to evaluate
  5. Connect nodes to the True output port (executed when condition is met)
  6. Connect nodes to the False output port (executed when condition is NOT met)
  7. Run the workflow — only one path executes

Workflow Examples

Content-Based Routing (Standard)

Text Node (article analysis) → If/Else (contains “breaking news”) → True: Scene with urgent red styling → False: Scene with calm blue styling

Quality Check (AI)

Reference Image → Text Node (“Describe this image’s quality”) → AI If/Else (condition: “Is this image high quality and suitable for a professional presentation?“) → True: proceed to Scene → False: route to Upscaler for enhancement first

Language Detection (AI)

Text Node (user input) → AI If/Else (“Is this text written in English?“) → True: proceed normally → False: route through Text Node for translation first

Tips & Best Practices

  • Skipped branches cost nothing — only the executed path consumes credits
  • Use Standard If/Else for simple, exact conditions (contains, equals, empty checks)
  • Use AI If/Else for subjective judgments (“Is this appropriate?”, “Is the quality good?“)
  • You can nest If/Else nodes for complex decision trees (If/Else → If/Else on the True path)
  • The AI If/Else uses one text generation call — choose a fast, free model (Gemini 3 Flash) for low-cost routing
  • Connect different node types to each branch — True path might create a video, False path might generate an image

Troubleshooting

Both paths seem to execute

Check your connections. If nodes after the If/Else have OTHER inputs that aren’t gated by the If/Else, those nodes may execute independently.

Condition always evaluates to True (or always False)

  • Standard: Check operator and comparison value. “Contains” is case-sensitive.
  • AI: Make your condition prompt more specific and unambiguous.

AI If/Else is slow

The AI must read the input and evaluate — this takes a few seconds. Use Gemini 3 Flash for fastest evaluation.

Skipped nodes show errors

Skipped nodes in the non-taken branch should show as “skipped”, not “error”. If they show errors, check that they’re connected ONLY to the If/Else output (not to other sources that might trigger them).

See Also

  • Text Nodes — Generate text for condition evaluation
  • Start Node — Beginning of workflows with branching
  • Scenes — Different video styles per branch
  • Reference Images — Different image generation per branch