OpenAI says Higgsfield is adding longer videos, more precise visuals, and a path toward voice filmmaking through three named models.
Higgsfield says it is expanding its AI video tools with GPT-6 Astra, GPT-Image-2.5, and GPT-Live, according to an OpenAI announcement published October 13.
The announcement describes GPT-6 Astra as enabling longer videos. It links GPT-Image-2.5 to more precise visuals, which could give creators tighter control over what appears on screen.
GPT-Live is described as offering a path to voice filmmaking. That phrase signals a direction for the product, but the announcement does not explain the specific feature, workflow, or controls involved.
For creators, the appeal is straightforward: longer output and more precise images could support work beyond short video experiments. The voice-filmmaking reference suggests Higgsfield is also exploring a different way to create video, though the practical benefit remains unclear until the company explains how it works.
This remains an announcement rather than an independent performance test. The available information does not say whether the capabilities are broadly available, limited to testing, or deployed at scale. It also does not provide launch timing, pricing, access rules, benchmark results, or comparisons with other video systems.
The useful next step for creators is to check whether Higgsfield has enabled these capabilities for their account and what the tools actually support. Until then, the announcement describes potential changes to video length, visual precision, and voice filmmaking—not a confirmed buying option.