Huffyuv's algorithm is similar to that of lossless JPEG, in that it predicts each sample and then Huffman-encodes the error.

The original implementation was written for Windows by Ben Rudiak-Gould and published under the terms of the GPL.

Huffyuv 2.1.1 with CCESP patch 0.2.5 was released to address problems particularly for compatibility with Cinema Craft Encoder.

There is a currently developed fork of the code named Lagarith which offers better compression at the cost of greatly reduced speed on uniprocessor systems.

FFmpeg also has its own FFmpeg-specific (incompatible with the original) version of HuffYUV which supports a wider variety of colorspaces, including chroma-quartered and high bit depth color spaces up to 48bpp.