Skip to main content

Introduction

fal.ai is a model-hosting platform, and on the AI Settings page one fal.ai key drives two Video Studio modelsKling 3.0 Pro (Kuaishou) and Seedance 2.0 (ByteDance). Both run through the fal.ai queue API.
One key, two models. Although Kling and Seedance come from different model makers, both are served via fal.ai here, so they share a single key. You won’t find separate “Kling” or “Seedance” cards in AI Settings — configure them both under the fal.ai vendor card.

How keys are stored

All AI provider keys live in the AI Settings screen and are stored encrypted in the database. There are no .env edits — keys are injected into the drivers at runtime. Users never supply their own key; every generation runs on the admin’s key.
A model only becomes available to users when both are true: the fal.ai API key is saved and the model is enabled in the vendor modal.

Before you start

  • Admin access to your MagicAds platform.
  • A fal.ai account with billing/credits set up.

Part 1 — Get your fal.ai API key

1

Open the fal.ai dashboard

Go to fal.ai/dashboard/keys and sign in.
2

Add billing / credits

Make sure your fal.ai account has credits or billing set up so your key can make calls. fal.ai charges you directly for usage — separate from your MagicAds platform credits.
3

Create an API key

Create a new key and copy it.
The key is shown only once. Copy it immediately and store it somewhere safe.
4

Confirm model access

Your fal.ai account just needs credits — both Kling and Seedance run on the shared key.

Part 2 — Add the key in MagicAds

1

Open AI Settings

In your platform, go to Admin → AI Settings.
2

Open the fal.ai vendor

Click the fal.ai (Seedance, Kling) vendor card to open its configuration modal.
3

Paste the key

Paste your API key into the API key field. Leaving it blank keeps the existing key unchanged.
4

Enable the models you want

Toggle on Kling 3.0 Pro and/or Seedance 2.0, and set each model’s credit cost (what a user is charged per generation).
5

Save

Click Save. The vendor card shows as connected and the enabled models appear in the Video Studio.
There is no “test connection” button. To verify, enable a model and run a quick generation in the Video Studio.

Part 3 — Supported models

Video Studio

ModelModel IDTierDefault creditsDurationsAudioBest for
Kling 3.0 Profal-ai/kling-video/v3/proPremium104, 5, 6, 8, 10, 12, 15sYesCinematic-quality leader. Best at human motion, orbit/360° product shots, and complex camera moves, with native audio and multi-shot support.
Seedance 2.0bytedance/seedance-2.0Budget44, 5, 6, 8, 10, 12, 15sYesByteDance’s second-gen video model with native synchronized audio, multi-shot editing, and director-level camera control. Great for high-energy promos and motion-graphic content.
fal.ai provides video models only in AI Settings — no copy or image.
Credit costs shown are the seeded defaults. You can change each model’s cost in the vendor modal. These are what your platform charges users in credits — not fal.ai’s prices.

Part 4 — Also used by the UGC Factory plugin

The UGC Factory plugin also runs on a fal.ai key (for VEED Fabric 1.0 talking videos, ElevenLabs TTS, and actor generation). UGC Factory has its own fal.ai key field on its plugin settings page, but if that field is left blank it falls back to this shared fal.ai key from AI Settings. So configuring fal.ai here can power both the Video Studio (Kling/Seedance) and UGC Factory.

Part 5 — Make a fal.ai model the default video model (optional)

In the global section of AI Settings you can set the default video model to Kling 3.0 Pro or Seedance 2.0 so it’s pre-selected for users. You can also toggle the Video Studio on/off and control free-tier access there.

Troubleshooting

SymptomLikely causeFix
Kling/Seedance don’t appear in Video StudioKey missing or models disabledAdd the key in the fal.ai vendor modal and enable the models, then Save.
”The payload is invalid.” on generateSaved key can’t be decrypted (e.g. after an environment change)Re-enter the fal.ai key in AI Settings and Save to refresh it.
”fal.ai is rate-limiting requests”Too many concurrent jobsWait and retry; fal.ai throttles bursts.
fal.ai API errors / 401Invalid key or no billingRe-check the key and billing in the fal.ai dashboard.
Looking for a separate Kling or Seedance cardThey share the fal.ai keyBoth are configured under the fal.ai (Seedance, Kling) vendor card.
Keys are stored encrypted and applied at runtime. No .env changes or commands are needed after saving.