Creatify Aurora is now on fal.ai

Aurora is now available directly on fal.ai's platform.

Developers can access the same Aurora model that powers video ads for Comcast, Alibaba, and thousands of brands on Creatify - but now through fal's inference API without running it through Creatify's interface.

What fal.ai does

They've built the fastest inference for models like SDXL and Whisper. Now they're offering Aurora as part of their avatar model catalog.

fal.ai runs fast inference for generative AI models. Their platform serves hundreds of millions of users and specializes in making generative models accessible to developers at scale.

Why this is huge

Until now, Aurora was only available through Creatify's platform, and recently via ElevenLabs. That works great if you're creating video ads through our URL-to-Video tool, script-to-video, or AI Avatar features.

But developers building custom applications, agencies with specific workflow requirements, or technical teams integrating avatar generation into their own products needed direct API access.

fal.ai provides that access. You get Aurora's diffusion transformer architecture, ultra-realistic lip-sync, and full-body expressiveness through a standard API endpoint.

How to use Aurora on fal

The integration is straightforward. You need three inputs:

• image_url - your avatar image

• audio_url - speech or audio file

• prompt - optional visual direction (lighting, framing, background)

Here's the minimal setup:

const result = await fal.subscribe("fal-ai/creatify/aurora", {
  input: {
    image_url: "https://example.com/avatar.png",
    audio_url: "https://example.com/speech.mp3",
    prompt: "4K studio interview, medium close-up, soft key-light",
  },
});
console.log(result.video.url); // Download URL for generated video

The response returns a video URL with your generated avatar.

Pricing is $0.10 per second at 480p or $0.14 per second at 720p. Seconds round up for billing.

Important prompting detail

Aurora prompts work differently than text-to-image models. You're not describing what the avatar says - that comes from your audio file.

Instead, prompts describe the visual context: lighting, framing, background, presentation style.

Bad prompt: "Professional video, good lighting, person talking"

Good prompt: "4K studio interview, medium close-up (shoulders-up crop). Solid light-grey backdrop, uniform soft key-light, no lighting change. Presenter faces lens, steady eye-contact. Hands remain below frame, body perfectly still."

fal published a comprehensive guide covering prompt architecture, common failures, and production-ready patterns. If you're building with Aurora on their platform, read their full prompt guide before generating your first video.

Who this is for

This integration serves developers who need:

• Direct API access for custom applications

• Integration into existing video pipelines

• Programmatic avatar generation at scale

• Control over technical parameters and workflows

If you're creating video ads from product URLs or using our platform's creative tools, keep using Creatify. The URL-to-Video workflow, AI Script Writer, and batch processing remain the fastest path from product to finished ad.

If you're building custom applications or need programmatic control over every parameter, Aurora on fal.ai gives you that flexibility.

Same Aurora model. More places to build.

You can use Aurora in fal by searching "Creatify" or "Aurora" in fal's model catalog to get started.

Try our latest Creatify Aurora model, and more than 1500 unique avatars for your marketing videos.