Skip to content
modelBridge Docs

Your First Generation

This guide takes you from zero to your first AI-generated result inside Premiere Pro. The whole process takes about 2 minutes.

Step 1: Select a Clip

Section titled “Step 1: Select a Clip”

Click on a video or image clip on your Premiere Pro timeline. modelBridge detects the selection and shows clip info in the panel — duration, resolution, and media type.

Step 2: Pick a Model

Section titled “Step 2: Pick a Model”

Use the model selector at the top of the panel. You can:

  • Browse your installed models
  • Search by name — “kling”, “wan”, “ltx”
  • Filter by category — text-to-video, image-to-video, etc.

For your first generation, try Kling Video v3 or Wan 2.6 — both are reliable image-to-video models with fast generation times.

Step 3: Write a Prompt

Section titled “Step 3: Write a Prompt”

Describe what you want the AI to generate. Be specific about motion, camera, and mood:

“Slow dolly forward through a misty forest at dawn, soft golden light filtering through the trees, cinematic”

More detail generally produces better results. See Prompt Tips for more guidance.

Step 4: Check the Cost

Section titled “Step 4: Check the Cost”

Look at the cost badge near the Generate button:

  • $X.XX · Estimated — a forecast from fal.ai’s published price for this model, applied to your settings; not a quote
  • ≈$X.XX · Learned — priced from your own previous runs of this model at these settings; still a forecast
  • $X.XX / 1K tokens · Rate — fal.ai bills this model per block of tokens; the count is set by the finished video, so no total can be given beforehand
  • From $X.XX — a floor, not a forecast: fal.ai’s lowest rate for this model; your settings can bill more
  • No price — fal.ai prices this model in a unit that only exists once the run has finished; there is nothing to multiply beforehand

The estimate updates live as you change parameters like duration and resolution.

Step 5: Generate

Section titled “Step 5: Generate”

Click Generate. The button transforms into a progress indicator:

  1. Submitting → 2. Queued → 3. Generating → 4. Downloading → 5. Importing

Step 6: See the Result

Section titled “Step 6: See the Result”

The result imports to your project automatically. A preview appears in the panel so you can evaluate it before using it in your edit.

From here you can:

  • Import to Timeline — places the result on your timeline automatically
  • Save to Project Bin — imports without placing on the timeline
  • Discard — removes the result

Keep Working While It Generates

Section titled “Keep Working While It Generates”

If a generation takes longer than expected, modelBridge hands it off to the background automatically. You can:

  • Continue editing your timeline
  • Switch to a different model and generate something else
  • Track progress in the Active Generations Panel
  • Get notified when the result is ready

Your edit never has to pause.

What to Try Next

Section titled “What to Try Next”