AIAny
AI Video2026
Icon for item

Veyra

Runs local video, capture-card feeds, and game streams through a real-time GPU enhancement pipeline with super resolution, denoising, HDR processing, and frame generation. Supports NVIDIA and AMD GPUs on Windows 11; some models and features depend on specific hardware or runtimes.

Introduction

Most video enhancement tools focus on files or a single playback path. The more unusual idea here is to put local media, capture cards, and game streaming through the same live processing pipeline—so enhancement can happen before the picture reaches the display, not only in an offline export. That makes it relevant to people trying to improve a console or PC signal as much as to ordinary video playback.

What Sets It Apart
  • One pipeline for live and file-based sources: The shared processing engine handles playback, capture, streaming, and List-mode export. This means effects and source handling can carry across workflows, though Node-mode graphs are not supported for offline export.
  • Layered, selectable enhancement: Super resolution, noise reduction, colour tools, HDR processing, and frame generation can be combined, with independent settings for multiple NR layers. The practical benefit is control over trade-offs; more layers and higher resolutions also raise GPU and VRAM demands.
  • A native GPU-oriented implementation: The Windows application uses DirectX 12 for video rendering and supports NVIDIA and AMD paths. Its README describes GPU-specific model options, not a universal feature set—compatibility and performance depend on the card, drivers, and available runtime assets.
  • Designed around real inputs: Capture-card, console, PC-streaming, and screen-capture sources make it useful beyond a conventional media player. The Xbox route is explicitly experimental, and hardware, network, HDR, and controller combinations still need individual validation.
Who It’s For

Great fit if you want to experiment with real-time enhancement of video or game-capture signals and are comfortable tuning GPU-dependent effects. Look elsewhere if you need a simple player, guaranteed support for every GPU or capture device, or a fully vendor-certified DLSS 5 integration: the project states that it is not a complete official DLSS 5 integration, and some runtime components are not distributed with the source.

Information

Categories

More Items

Hugging Face
AI Model2026

Generates video with synchronized stereo audio from text, keyframes, or reference media in 4–8 sampling steps. Pruning, low-rank factorization, and INT8/GGUF formats reduce runtime costs, though speed and hardware needs vary by setup.

Hugging Face
AI Video2026

Remakes existing footage from text instructions, changing characters, props, or scenes while preserving the original motion, camera, cuts, and sound. Use it in the Viggle app or run it on your own GPUs.

GitHub
AI Video2026

Lets coding agents edit local video and audio through 42 structured FFmpeg tools, then verify the output. The tools are also available through MCP, with no cloud service or API key required.