Model Search

Model Search lets you browse and add models from fal.ai’s full catalog without leaving Premiere Pro. Type a query, filter by category, and add any model with one click.

Searching

Type a query in the search field — “video upscaler”, “image to video”, “style transfer”, or a model name like “kling” or “wan”. Results appear as you type.

Each result card shows:

Model name — first row
Category badge — abbreviated labels (Img-to-Vid, Txt-to-Img, Vid-to-Vid, etc.)
License badge — green “commercial ✓” or amber “research” when available
NEW badge — orange badge for models added to fal.ai within the last 14 days
Description — third row, from the model’s API metadata

Filter Chips

Six category filter chips sit above the search results:

Chip	What It Shows
All	All models matching your search query
Image Gen	Text-to-image and image-to-image models
Video Gen	Text-to-video and image-to-video models
Upscale	Image upscaling and enhancement models
Inpainting	Inpainting and fill models
Extend Video	Video extension models

Some chips map to multiple API categories — “Video Gen” fetches both text-to-video and image-to-video results in parallel.

Commercial Filter

A Commercial filter chip works as an independent toggle. Combine it with any category to show only models with a commercial use license. Useful when you need to ensure generated content is cleared for client work.

The filter chip row scrolls horizontally when chips overflow the panel width.

Sorting

Four sort options are available:

Relevance — fal.ai’s default ranking (recommended for general searches)
A → Z — alphabetical
Z → A — reverse alphabetical
Newest — most recently updated models first

Sorting is applied to all loaded results.

Pagination

The initial search loads 15 results. Click Load more at the bottom to fetch the next page. Results accumulate — you can scroll through everything loaded so far.

For filtered chips that do client-side keyword matching (Upscale, Inpainting), the initial fetch loads 30 results to compensate for post-filter reduction.

Adding a Model

Click the +Add button on any search card. The plugin fetches the model’s OpenAPI schema and pricing data, then registers it in your installed models.

The button shows visual feedback through each state:

Add — default state
Adding… — shimmer animation while schema loads
Added — green flash, then the card fades to 50% opacity

After adding, the model appears in your installed models dropdown and is ready to use immediately.

How Schema Loading Works

When you add a model, modelBridge fetches its OpenAPI specification from fal.ai. This schema describes every input the model accepts — field types, constraints, defaults, and required fields. The plugin then:

Parses input fields — classifies each field as a slider, dropdown, text input, media upload, checkbox, etc.
Extracts constraints — minimum/maximum values, allowed formats, required fields
Detects media requirements — what types of input media the model needs (image, video, audio, or none)
Fetches pricing — queries fal.ai’s pricing API for cost estimation data
Caches everything — schema and parsed inputs are stored locally so future loads are instant

Schema caching uses a 5-minute TTL. Schemas are refreshed automatically when stale.

Troubleshooting

No results appear

Check your internet connection — search requires access to fal.ai’s API
Try a broader query — fal.ai’s search is keyword-based, not fuzzy
Try a different filter chip — your query might match a different category

Model fails to add

The model’s schema may be temporarily unavailable on fal.ai
Some models are gated and require access approval from the model creator — see the error message for details
Check the Active Generations Panel or error banner for specific error messages

Search is slow

modelBridge checks server health before each search. If the local server is recovering, there may be a brief delay
Large result sets with many categories load in parallel but may take a moment on slow connections