OpenAI has announced updates to Higgsfield, its AI video generation platform, significantly extending video length capabilities from one minute to over 10 minutes. According to OpenAI, the enhanced platform brings creators three major improvements designed to expand creative possibilities in AI-generated video content.
The updates center around new model integrations. GPT-6 Astra enables the extended video duration, allowing creators to produce substantially longer content than previously possible. GPT-Image-2.5 focuses on delivering more precise visual generation, addressing accuracy in the video creation process. Additionally, the platform introduces GPT-Live, which OpenAI describes as providing “a path to voice filmmaking,” suggesting voice-driven video creation capabilities.
According to the announcement, these enhancements position Higgsfield as a more comprehensive tool for creators working with AI-generated video content. The jump from one-minute to over 10-minute videos represents a tenfold increase in duration capacity, potentially opening new use cases for longer-form AI video content creation.