Higgsfield says its AI video workflow now produces videos longer than 10 minutes

Higgsfield says its AI production workflow can now create videos exceeding 10 minutes, compared with roughly one minute previously. CEO Alex Mashrabov described the change in an October 13 company case study published by OpenAI.

The workflow uses GPT-6 Astra to coordinate creative agents and external tools, including tools for 3D work. Mashrabov said those capabilities help preserve visual consistency across longer productions. Higgsfield uses the workflow to make instructional videos for its own channel.

The company also uses GPT-Image-2.5 for promotional images containing text and infographics. Mashrabov said these capabilities help smaller creative teams handle projects that previously required more production resources.

Higgsfield next wants creators to describe a film aloud while AI builds the visuals as they speak. The announcement presents GPT-Live-1 as part of that ambition, but gives no release date for the proposed experience.