Skip to main content

What are AI models and how to use them?

Updated today

AI models are the creative engines behind AI generation in Async.

When you create content in the Video Editor, you can use different models to generate images, videos, avatars, music, and sound-based content. Each model has different strengths, so the best one depends on what you want to make.

What are AI models?

AI models are the systems that power different types of content creation.

Some models are designed for:

  • image generation

  • video generation

  • avatar creation

  • music and sound generation

  • upscaling existing content

Different models can vary in style, speed, quality, realism, and control.

Where do I find AI models in Async?

When you open the Video Editor, AI Chat opens automatically and gives you different ways to start creating. From there, you can type a prompt, upload a photo or video, or click Explore models to browse your options.

What can you create with AI models?

With Async’s AI models, you can create:

  • Images

  • Videos

  • Audio & voice

  • Avatars

  • Music and sound effects

  • Upscaled image and video content

How to use AI models

  1. Open the Video Editor.

  2. The AI Chat will open automatically.

  3. Enter a prompt describing what you want to create, or upload a photo or video to work on.

    Chat
  4. If you’re not sure which model to use, click Explore models.

    Explore
  5. Choose the content type you want to create.

  6. Select the model that best fits your goal.

  7. Once you confirm everything, click Generate.

Which AI models do we offer?

Image generation

Our image generation models include:

  • FLUX.2

  • Kling-image/o1

  • Kling-image/o3

  • Seedream 5.0 Lite

  • Bytedance Seedream v4.5

  • Bytedance Seedream v3

  • Wan v2.6

  • Wan 2.5

  • Nano Banana Pro

  • Nano Banana 2

  • Nano Banana

  • Gemini 3 Pro Image

  • Gemini 2.5

  • Grok Imagine Image

  • Recraft 4 text-to-image

  • Recraft 3 image-to-image

  • Recraft 2 text-to-image

Image upscaling

For image upscaling, we offer Topaz upscale image, with options up to:

  • 24MP

  • 48MP

  • 96MP

  • 512MP

Avatars

Our avatar model includes:

  • HeyGen Avatar 4

  • Creatify Aurora

Video generation

Our video generation models include:

  • Hailuo-02 Pro

  • Hailuo 2.3

  • Hailuo 2.3 Fast

  • Kling Video v2.6

  • Kling Video v3

  • Kling Video o1

  • Kling Video o3

  • Bytedance Seedance 2.0

  • Bytedance Seedance v1.5 Pro

  • Bytedance Seedance 1

  • Sora 2

  • Veo 3

  • Veo 3.1

  • Wan v2.6

  • Wan flf2v

  • Wan 2.5

  • Grok Imagine Video

  • Creatify Aurora

  • Seedance 2.0

Video upscaling

For video upscaling, we offer Topaz upscale video, including:

  • up to 720p

  • 720p to 1080p

  • 1080p and above

Music and sound generation

Our music and sound models include:

  • Mirelo AI SFX v1.5

  • Mirelo AI SFX v1

  • AsyncFlow AI music generator

New AI models are being added constantly. Please click here for the full list.

Explore models

Different models are designed for different kinds of creative work. Some models are better for speed, some for realism, some for readable text, and some for keeping the same character or subject consistent across multiple assets.

Here are a few examples:

  • FLUX.2 is a strong option when you need visual consistency across multiple images.

  • Gemini Image is useful for images with readable text, like posters, signs, or labels.

  • Hailuo is a good choice for cinematic short-form video generation.

  • Veo 3 / 3.1 is useful when you want video and audio generated together.

A simple way to choose is to start with the type of content you want to create, then test a couple of models to see which result fits your project best.

💥Tips & Tricks💥

  • Start with the type of content you want to create first.

  • If you’re unsure, use Explore models to compare your options.

  • Start simple, then refine your prompt step by step.

  • Match the model to the task.

  • Try more than one model if you want to compare results.

Did this answer your question?