{"name":"vmaf","repo":"Netflix/vmaf","url":"https://midorreal.com/project/netflix-vmaf","repository":"https://github.com/Netflix/vmaf","description":"Perceptual video quality assessment based on multi-method fusion.","facts":{"language":{"value":"C","source":"https://github.com/Netflix/vmaf","checked":"22 Nov 2026"},"last_release":{"value":"v3.2.1, 14 Sep 2026","source":"https://github.com/Netflix/vmaf/releases","checked":"22 Nov 2026"},"releases":{"value":"26","source":"https://github.com/Netflix/vmaf/releases","checked":"19 Nov 2026"},"contributors":{"value":"121","source":"https://github.com/Netflix/vmaf/graphs/contributors","checked":"19 Nov 2026"},"archived":{"value":"no","source":"https://github.com/Netflix/vmaf","checked":"22 Nov 2026"}},"stars":5623.0,"writeup":{"what_it_is":"VMAF is a perceptual video quality assessment algorithm developed by Netflix. The package includes a C library (libvmaf), a Python library for wrapping and training custom VMAF models, and a command-line tool. It also includes implementations of PSNR, PSNR-HVS, SSIM, MS-SSIM and CIEDE2000.","audience":"Video encoding and codec evaluation engineers, and developers who measure video quality.","claims":[{"kind":"specific","claim":"Includes a stand-alone C library and a wrapping Python library","excerpt":"This software package includes a stand-alone C library `libvmaf` and its wrapping Python library.","status":"not_checked"},{"kind":"specific","claim":"Includes implementations of PSNR, PSNR-HVS, SSIM, MS-SSIM and CIEDE2000","excerpt":"implementations of several other metrics: PSNR, PSNR-HVS, SSIM, MS-SSIM and CIEDE2000.","status":"not_checked"},{"kind":"specific","claim":"Available as a filter in FFmpeg when configured with --enable-libvmaf","excerpt":"VMAF is now included as a filter in FFmpeg, and can be configured using: `./configure --enable-libvmaf`.","status":"not_checked"},{"kind":"specific","claim":"The v2.0.0 release has a fixed-point, x86 SIMD-optimized implementation that achieves a 2x speed up over the previous floating-point version","excerpt":"It has a new fixed-point and x86 SIMD-optimized (AVX2, AVX-512) implementation that achieves 2x speed up","status":"not_checked"},{"kind":"broad","claim":"Provides a command-line tool, a C library and a Python library for different uses","excerpt":"The command-line tool [`vmaf`](libvmaf/tools/README.md) provides a complete algorithm implementation","status":"too_broad"}],"alternatives":[],"written_by":"AI, from the project's README","date":"2026-11-02"},"why_now":null,"owner_supplied":null,"score":{"verdict":"mid","hype":27,"reality":36,"gap":-9,"momentum":null,"confidence":60,"version":"score-2.0","rule":"verdict-2.0","calculated_at":"2026-11-22T07:00:00+00:00","verdict_source":"rule"},"early_read":null,"changes":[{"date":"2026-09-14T16:03:17+00:00","text":"Latest release v3.2.1","source":"https://github.com/Netflix/vmaf/releases"},{"date":"2016-06-17T23:26:53+00:00","text":"First release","source":"https://github.com/Netflix/vmaf/releases"},{"date":"2016-02-08T18:41:38+00:00","text":"Repository created","source":"https://github.com/Netflix/vmaf"}],"cite":"vmaf is a GitHub project at https://github.com/Netflix/vmaf written mainly in C, described there as \"Perceptual video quality assessment based on multi-method fusion\".","attribution":"Data from Mid or Real, https://midorreal.com/project/netflix-vmaf, checked 22 November 2026. Attribution with a link is required.","still_real":null}