From 85967a43f6c04e0637de636d4ea2a30dd813d6f2 Mon Sep 17 00:00:00 2001 From: Adhitya Nadooli <127269031+adhityanadooli@users.noreply.github.com> Date: Wed, 1 Apr 2026 13:42:04 -0400 Subject: [PATCH 1/2] Add Vivid; Remove enhancr (unmaintained) from FAQ Removed enhancr entry from the FAQ section. Vivid successes Enhancr --- docs/faq.md | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/docs/faq.md b/docs/faq.md index 2b74de64..bc0ebf25 100644 --- a/docs/faq.md +++ b/docs/faq.md @@ -26,14 +26,15 @@ There are many programs that can perform AI upscaling, though not all are able t - [Download](https://chainner.app/download) | [GitHub](https://github.com/chaiNNer-org/chaiNNer) | [Website](https://chainner.app/) - chaiNNer is a node-based image processing GUI that can also be used for image upscaling. It has the most support for models listed on OpenModelDB and can be run with PyTorch (CUDA), ONNX (CUDA), and NCNN (AMD/Intel). Being node based, it allows you to process your images with a lot more control, at the cost of being a bit more complex. Besides upscaling, chaiNNer also has a variety of other use cases and is a very versatile program. +- **Vivid** (closed-source | paid) + - [GitHub (Extension SDK)](https://github.com/dusklaboratory/vivid-core) + - Desktop AI video processing platform focused on upscaling, frame interpolation, and restoration. It supports multiple runtimes including TensorRT, NCNN, DirectML, CoreML, and ONNX Runtime, enabling efficient processing across NVIDIA, AMD, Intel, and Apple Silicon hardware. Built-in model hub and custom model support! + - Unlike traditional tools, Vivid is built around pipelines and queue-based workflows, allowing users to chain operations like upscaling, interpolation, and restoration into a single process. It also supports optional cloud acceleration for significantly faster processing when local hardware is insufficient. + - **AnimeJaNaiConverterGui** (open-source | free) - [GitHub](https://github.com/the-database/AnimeJaNaiConverterGui) - Allows for fast video upscaling utilizing TensorRT (On Nvidia cards), DirectML, or NCNN within a clean GUI. Only supports ONNX. Use chaiNNer to convert PyTorch models to ONNX for usage. -- **enhancr** (open-source | paid) - - [GitHub](https://github.com/mafiosnik777/enhancr) - - Meant for fast video frame upscaling and interpolation, enhancr takes advantage of Nvidia's TensorRT to provide fast processing on video. It supports a variety of models, including many of the custom ESRGAN ones from OpenModelDB. - - **VSGAN-TensorRT-docker** (open-source | free) - [GitHub](https://github.com/styler00dollar/VSGAN-tensorrt-docker) - A more complicated but free alternative for fast video upscaling with TensorRT. Supports an insane amount of models, but requires manual parameter selection. From 41e53448f00effccfe0912cd1ccd89fafa59eefe Mon Sep 17 00:00:00 2001 From: Adhitya Nadooli <127269031+adhityanadooli@users.noreply.github.com> Date: Wed, 1 Apr 2026 13:43:49 -0400 Subject: [PATCH 2/2] Fix Vivid link in FAQ Updated the Vivid entry with a new website link. --- docs/faq.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/faq.md b/docs/faq.md index bc0ebf25..51775c06 100644 --- a/docs/faq.md +++ b/docs/faq.md @@ -27,7 +27,7 @@ There are many programs that can perform AI upscaling, though not all are able t - chaiNNer is a node-based image processing GUI that can also be used for image upscaling. It has the most support for models listed on OpenModelDB and can be run with PyTorch (CUDA), ONNX (CUDA), and NCNN (AMD/Intel). Being node based, it allows you to process your images with a lot more control, at the cost of being a bit more complex. Besides upscaling, chaiNNer also has a variety of other use cases and is a very versatile program. - **Vivid** (closed-source | paid) - - [GitHub (Extension SDK)](https://github.com/dusklaboratory/vivid-core) + - [Website](https://vividenhance.com/) | [GitHub (Extension SDK)](https://github.com/dusklaboratory/vivid-core) - Desktop AI video processing platform focused on upscaling, frame interpolation, and restoration. It supports multiple runtimes including TensorRT, NCNN, DirectML, CoreML, and ONNX Runtime, enabling efficient processing across NVIDIA, AMD, Intel, and Apple Silicon hardware. Built-in model hub and custom model support! - Unlike traditional tools, Vivid is built around pipelines and queue-based workflows, allowing users to chain operations like upscaling, interpolation, and restoration into a single process. It also supports optional cloud acceleration for significantly faster processing when local hardware is insufficient.