Skip to content

Dual Mode

Dual Mode runs two AI models at the same time from a single Generate click. Instead of trying one model, waiting, then trying another, you get both results at once — ready for comparison.

  1. Select a primary model from the dropdown
  2. Enable Dual Mode using the toggle
  3. Select a secondary model from the second dropdown
  4. Click Generate — both models run in parallel using the same prompt and source media

Both results appear as selectable cards. Click one to preview, import it, and the other card auto-selects so you can decide on the second result immediately.

Before generation starts, modelBridge compares both models’ inputs and identifies shared fields. Fields that represent the same concept but use different names — like prompt vs multi_prompt — are matched automatically so you only configure once.

If the secondary model has a required field with no match in the primary model’s form, generation is blocked with a specific warning.

The cost badge shows the combined estimated cost for both models. The display adapts to available pricing data:

  • Both in USD → $X.XX · 2 models with per-model breakdown
  • Mixed types → $X.XX + From $X.XX · 2 models
  • One or both unavailable → shown honestly

A confirmation dialog appears when the combined cost exceeds $5.00.

Both models use the same source clip. When importing:

  • First result imports with Replace on Timeline — the source clip is replaced
  • Second result automatically falls back to Insert at Playhead — since the source clip no longer exists
  • No conflicts — you never have to think about which result replaced what
  • Comparing quality — test two models on the same prompt to see which works better for your use case
  • Speed vs quality — run a fast model alongside a high-quality model to see if the difference matters
  • Comparing voices — generate the same script with two TTS models
  • Client presentations — produce two options without doubling your workflow
  • Both models always use the same source clip
  • Cost history logs dual generations as two independent entries
  • Parameters unique to the secondary model use that model’s defaults