Introducing Pika H3 Flex

Pika has introduced Pika H3 Flex, a post-trained MiniMax H3 text-to-video model designed to give creative professionals faster generation speeds and deeper control over post-production. The model generates up to 15-second clips with synchronized audio, accepts up to 12 reference files combining images, videos, and audio, and supports 4K upscaling. Additionally, Pika is releasing open weights for the generator adapters and inference code. Open article →