Pika offers Pikaformance, an AI model that takes a still image of a face and an audio track as input, and generates a video of the image speaking, singing, or reacting in sync with the audio. It is available to registered users on the Pika web platform and iOS app.

Pikaformance is described by Pika on its login page as an 'audio-driven performance model' that produces 'hyper-real expressions, synced to any sound' at near real-time generation speed. It accepts a portrait image and an audio file (speech, music, singing, rapping, sound effects) as input, and outputs an animated video with lip movements, micro-expressions, and eye movements matched to the audio. It is priced at 3 credits per second of audio on paid plans. The model was confirmed available on the Pika web platform as of 2025.