Understand the Topaz Video AI user interface.

Drag and drop a file onto the interface or select 

 to locate the file(s) you wish to enhance and upscale. 

The video library will display the files you are working with. You can add or remove files at any time with ease. 

Files will be displayed in the order that they are imported. 

When working with a large batch, use the scroll bar or mouse wheel to scan through the files.

To adjust the segments in Topaz Video AI drag the horizontal line that separates the in-app video player from the library up or down. 

Add additional files by dragging and dropping or using the 

A file count will be displayed to keep track of the number of files loaded. 

To remove a file, select the triple dot menu and  

Use the playback and video edit toolbar to control playback on previews, crop, trim, and more.

Move the playhead to scan through the file and determine which section of the file you wish to preview. 

When you move the playhead, you'll see the length of the file and specific timestamp.

Use the [ ] buttons to select the range of the input video that will be processed during exports. Keyboard shortcuts I and O cam also be used to set this range.

Choose from one of the default presets or use the + icon to create your own. Presets can be edited from the 

Use this menu to review the input file's resolution and frame rate and to set the exported file's resolution and frame rate. 

Topaz Video AI offers many filters that have been trained for specific use cases. Maximize your experience by learning more about 

Enable one filter or two filters. Hovering over the filter sections will reveal tooltips to familiarize yourself with the models. 

Select the encoder, container, and audio selection for your exports. 

Previewing will determine if the best filter(s) have been selected for your project.

Select a preview range from 1 second to 15 seconds. 

Press Preview to preview your selections. 

Exporting your file is the final step in the Topaz Video AI workflow. 

 to save the file to the default output location.

 (right click) to save the file to a new location output location.

User Interface

Learn how to import, preview, and export. 

There are two ways to bring file(s) into the app. 

Open the app and select Browse. From here, select the video files.

Drag and drop files directly into the app.

Video AI supports a wide variety of video codecs and containers. For a full list of compatible input files, 

please refer to this chart for libavcodec

Import image sequences using the same steps listed above. 

When using an image sequence make sure that:

The "Video In" menu will display the input file's resolution, aspect ratio, and frame rate.

The "Video Out" menu will allow you to alter the output file's resolution and frame rate.

If you are using a video file with a square aspect ratio an intermediate menu option will allow you to keep the original pixel type or square pixels.

Use the toggle switch to enable filter(s). We recommend using one to two per file.

The Output Settings menu allows you to alter the export encoder, container, and audio settings.

To generate a preview, select the resolution, filters, and output settings and press the 

You can select up to 15 seconds to preview and you can adjust the view and zoom to inspect the results.

You can generate as many previews as you need. 

The progress of each preview will be shown in the Output library. 

A purple progress bar and an ETA will appear for each file. 

Once the preview is ready, you will see a green checkmark.

To view the Preview using your default video player, double-click on the preview or use the triple dot menu to open an external player.

Once you have inspected the preview, you are ready to export. 

By default, the exported file will be saved to the input file's location. 

 if you wish to change the save location. As a note, 

In Video AI v5, exports can be paused and resumed even after closing the app or rebooting your system. 

Paused exports will be saved in multiple segments and joined together once processing completes. 

The following settings must be enabled in Preferences -> Export for pause/resume to be available in the app menu: 

Importing, Previewing & Exporting

Choose a premade preset or make your own. Learn the use case for each Preset, and how to create, modify and delete Presets. 

Open Topaz Video AI and locate the Preset filter at the top right-hand corner of the app. 

After choosing a Preset, you can continue to add or change the model selection. 

 from the dropdown menu or select the filters you wish to combine and then select 

 to slow down the playback speed of your file.

For speed changes above 4x, we recommend using Apollo. 

This will reduce the output resolution, however, this setting is recommended for most stabilization needs. 

We recommend using this model when altering the frame rate of files that are not in slow motion. 

 model to deinterlace and upscale your interlaced file.  By default, 

The output frame rate will be doubled. If you wish to retain the original frame rate or change it, select the desired frame rate from the 

 to upscale to a resolution of 4K (3840x2160). 

You can change the Enhancement model to reflect your unique use case. 

 to upscale to a resolution of 4K (3840x2160) and set the frame rate to 60 FPS.

 to upscale to a resolution of HD (1920x1080). 

1. Select the preset you wish to edit. 
2. Select Edit to 

Presets

You can quickly set In and Out points, trim clips in the Timeline, crop, and more.

To trim, use the [ ] buttons on the left side of the app's timeline. 

The in-point in the sequence where the clip begins and the out-point in the sequence where the clip ends. 

If at any point you wish to alter the trim selection, press 

To set the trim by frame number, select frame number instead of timecode in the 

Allows for more precise control over the range of the selected video. 

This setting will also change the length of Previews in Video AI to be displayed in frames instead of seconds.

To crop, locate the crop tool which is denoted by a crop icon on the left-hand side of the app. 

Choose an aspect ratio or set a custom height/ width/ x/ y axis or use the crop handles to adjust the selection. 

Trim and Crop 

Learn about the different encoders and containers that Topaz Video AI offers.

Changing the encoder selection converts information from one format to another for standardization, speed, and/or compression.

Choose from Topaz Video AI's 14 encoders. 

Some encoding options include a Dynamic mode, which will adapt the bitrate of the output video to match the visual complexity of the scene being processed. 

When available, most users should try the High or Medium setting for Dynamic mode before trying some of the constant bitrate settings.

Apple ProRes codecs take full advantage of multicore processing and feature fast, reduced-resolution decoding modes. These codecs are intended for use with editing software and create large output files, especially with high-resolution videos. 

ProRes codecs support SD, HD, 2K, 4K, and 5K at full resolution. 

The lowest-bitrate variation of the ProRes codec, intended for use in editing software as a smaller version of larger files like ProRes HQ.

Great for environments where storage capacity and data rate are limited.

Highly compressed with roughly 70 percent of the data rate and 30 percent smaller file sizes.

The original implementation of the ProRes codec, and the standard conversion format for Final Cut Pro X

Higher-data-rate version of Apple ProRes 422 that preserves visual quality at the same high level as Apple ProRes 4444 for 4:2:2 sources.

The highest-quality version of Apple ProRes for 4:4:4:4 sources.

Very high data rate to preserve the detail in high-dynamic ranges.

Can be used on macOS 12.0 and later with TVAI.

Advanced Video Coding (AVC) is one of the most common video compression standards in use today.

Most common codec for web and mobile video.

Will create slightly larger files than H265 encoders. 

Can be played in 99.9% of web browsers, mobile devices (including the iPhone/iPad, Android), and TVs.

High-Efficiency Video Encoding (HEVC) is an encoder that when compressed to the same bitrate as H.264, can offer higher-quality visual results. 

HEVC makes 4K videos easier to download and stream in the best quality possible.

The Main profile supports 8 bits per sample.

More advanced encoding setting within H265 optimized for high-detail videos

The Main profile supports 10 bits per sample.

Similar to H265 when it comes to quality and compression. Recommended for embedding in webpages.

Highly-efficient codec supported on NVIDIA RTX 4000 series and Intel ARC GPUs. May not play smoothly on all devices as AV1 is more system intensive than other codecs during playback.

Lossless compressed codec intended for use in archival workflows where large file sizes are not an issue. Supports 8, 10, and 12-bit output.

You can export your video file as an image sequence. An image sequence is a set of sequentially numbered frames that can be imported to a video editing program for further editing.

Use TIFF when you need a high-quality sequence and have the storage space to accommodate large files.

Use PNG when you need a high-quality image sequence. One of the best-supported uncompressed formats. Uses Lossless compression to save disk space without discarding any image information.

Encoders and Containers

Topaz Video AI allows you to customize various settings that can improve performance and the overall user experience. 

To access this menu:

Adjusting the processing preferences will control which hardware Topaz Video AI will use for:

How many resources Topaz Video AI is using.

The number of parallel tasks that are being handled.

The recommended number of tasks that can be processed at once is determined by your computer hardware. We do not recommend selecting a higher number of parallel processes as this setting may lead to instability. 

Determine which computer hardware is used to process, preview, and export tasks. 

The options displayed in the dropdown may differ based on your computer. 


Note: Apple Silicon processors will not have this option available, and will always use a combination of the device's CPU, GPU, and Neural Engine.

Specify the amount of system memory to allocate.

High resolution videos require a large amount of video memory and will benefit from a higher maximum memory usage limit. However, it's also possible to exceed machine resources and cause errors, which will require lowering the slider. 

Increasing max memory usage may impact the performance of other running applications.

Select the number of tasks to run in parallel. 

The maximum number of tasks is based on the hardware specifications of your system and will be limited in the dropdown to avoid system instability.

Enable this option to reduce the consumption of machine resources in order to conserve power.

Adjust the directories preferences to control where the models, temporary preview files, and exports are saved on your machine. 

This folder contains the AI models and resource files required by Topaz Video AI.

To alter the selection, click on the folder icon.

This folder contains generated preview files and other temporary files.

To alter the selection, click on the folder icon. 

This is the folder where all exported files will be saved. The input folder 

is the default location for exported files.

This setting controls the default location for preview files and project files (saved as ".tvai")

Adjust the output preferences to control the default encoder/ container, audio setting, and video export preview experience. 

Select the default video encoder to use when exporting files. Choose between video exports or an image sequence output. 

This selection will change based on the encoder selected. For more information on the profile, please visit the 

Select the default bitrate for exported video files.

Auto allows the selected encoder to choose the optimal bitrate.

Larger bitrate values will result in a larger file size of the exported video file.

This setting is only available when using h.264 or h.265 encoders.

We recommend leaving the audio mode set to the default selection of "copy" for most cases. Use "convert" when necessary for a certain output format. Selecting "none" will remove the audio stream. 

If it is necessary to choose "convert" when "copy" is selected, the app will notify you.

Choose the default file format for exported video files.

This toggle will allow you to control whether or not a preview is displayed in the in-app player. Disabling live preview will improve processing speed in most cases. 

This section of the preferences controls how much memory the experience uses. 

This setting determined where the preview files are stored on your machine. 

Alter this menu option to change the export preview experience.

With Preview: Temp files will be created in the export location while processing.

Without Preview: No temp files will be created and no recovery in case of failure.

Use Temp Directory (with preview): A temp folder will be used to store the intermediate files.

Keep Extra Headers (with preview): No temp files will be created and files may have issues with some editors/players.

Adjust the application preferences to control the preview experience, processing speed display, default preset, and the collection of anonymous data. 

Select the default view mode when initiating a preview task. 

Single will display the entire processed clip in the preview area. 

Side by Side will display the unprocessed clip in the left half of the preview area and the processed clip in the right half of the preview area. 

Split will display both the unprocessed and processed clips in the preview area and split them with a vertical bar that can be dragged left to display more of the processed video and dragged right to display more of the unprocessed video.

Preferences 

Advanced Functions

Filters

Dive into the details of various functions and features of Topaz Video AI. Learn different elements of the app and best practices for how to use them.