BioComputing launches Mischief 1 image-to-video model, 5x faster than its base
Article timeline
The articles in this story. Times are ET.
Rohan PaulXBiological Computing Co's Mischief 1 video model claims 5x Wan 2.2 speedAIBiological Computing Co says its Mischief 1 image-to-video model adds a layer based on findings from living human brain cells on electrodes to Wan 2.2. The layer adds less than 0.1% to Wan 2.2's parameters, and the company claims it matches Wan 2.2's quality at 5x the speed. Mischief 1 is available through an API at 0.5 cents per second, according to Alex Ksendzovsky.
elvisXBioComputing's Mischief 1 video model runs at least 5x faster than its baseAIBioComputing has spent four years studying how living neurons compute on electrodes, and its first product, Mischief 1, is a video model at least 5x faster than its base model. According to BioComputing's founder, Mischief 1 costs 0.5 cents per second of generated video and is the cheapest image-to-video model on fal and OpenRouter. Elvis Saravia says the research could improve AI efficiency, citing the brain's roughly 20 watts of power use.
SantiagoXMischief 1 image-to-video model is 5x faster and 80% cheaperAIThis is a great model for exploring ideas quickly and cheaply. It's an image-to-video generation model. 5x faster and 80% lower inference cost than its base open-weight model. • Input: a starting image and a prompt describing what you want • Output: a 720p video • Advantages: Fast generation, low cost. Easy to iterate. • Disadvantages: Output doesn't have production quality.