Video-native multimodal model

Pegasus 1.6 is built to understand video captured from a first-person perspective, a format increasingly common with wearable cameras and action-cam content. The release was announced through the company's official website and associated media assets, where the model's name and version were presented in its published metadata.

The model forms part of Twelve Labs' line of video-understanding AI, which focuses on letting software interpret video content through natural language rather than treating it as a sequence of disconnected frames.

Unclear details

Details such as benchmark results, pricing, availability through application programming interfaces, and specific performance comparisons with earlier versions were not included in the material reviewed. The company has not been contacted for additional comment, and independent third-party testing of the model's capabilities has not yet been reported.

Twelve Labs has previously attracted attention in the video AI space, and the new release continues its focus on video-native models rather than adapting text-first architectures to video tasks.

The claim that Twelve Labs released the first-person video analysis AI Pegasus 1.6 is, on balance, Mostly True.