av1an

Av1an is a cross-platform command-line video encoding framework written in Rust. Its main purpose is to speed up video encoding and improve CPU utilization by splitting input into chunks and running multiple encoder processes in parallel. Key capabilities described in the README: - Hyper-scalable video encoding through parallel encoder processes. - Target Quality mode, which uses quality metrics to control the encoder's rate control and reach a desired video quality level. - VapourSynth script support as input. - Cancel and resume encoding without losing progress. - Minimal command-line interface. - Docker images available. - Runs on Windows, Linux and macOS. Supported encoders (at least one is required): - aomenc (AV1) - SvtAv1EncApp (AV1) - rav1e (AV1) - vpxenc (VP8 and VP9) - x264 (H.264/AVC) - x265 (H.265/HEVC) The README notes that the executable encoder must be installed, not just the corresponding library. Installation options include package managers (Arch Linux/Manjaro via pacman, and Cargo via cargo install av1an), prebuilt Windows binaries in releases, a nightly build of master, and prebuilt Docker images. Manual installation requires FFmpeg, VapourSynth and at least one supported encoder. Optional components mentioned for better chunking or quality measurement include L-SMASH, DGDecNV, FFMS2, BestSource, mkvmerge, VMAF, XPSNR, Vapoursynth-HIP, Vapoursynth-Zig Image Process and vapoursynth-julek-plugin. These support chunking accuracy and Target Quality metrics such as VMAF, XPSNR, SSIMULACRA2 and Butteraugli. Basic usage examples show encoding an input file to an output file, or using a VapourSynth script with custom encoder parameters, worker count, target quality, audio encoding options and logging. Full documentation is available on the project's documentation site, and help is offered through Discord or GitHub issues.