Higgsfield extends AI video beyond 10 minutes
Higgsfield announced a workflow that uses GPT‑6 Astra for longer narratives, GPT‑Image‑2.5 for visual precision, and GPT‑Live as a path toward filmmaking with voice.
AT A GLANCE
Key points
- Video duration extends beyond 10 minutes
- GPT‑6 Astra structures longer narratives
- GPT‑Image‑2.5 supports visual precision
- GPT‑Live opens a path to voice filmmaking
What is changing?
OpenAI published a new Higgsfield workflow on October 13 that moves AI video production from one-minute clips to narratives longer than 10 minutes. The announcement highlights GPT‑6 Astra for building long-form structure and GPT‑Image‑2.5 for frame and scene precision.
Higgsfield also presents a path toward filmmaking with voice through GPT‑Live. The approach aims to handle visual continuity, scene sequencing, and the audio layer in one creative workflow instead of generating a single clip.
Visual assets on the official page were inspected but not copied because redistribution permission could not be verified. The “over 10 minutes” capability is a provider claim; teams should measure character and location consistency, editing control, audio synchronization, and cost in their own projects.
Explore the original sources
Source date: 13 October 2026.
Published by AI Radar: 14 Oct 2026, 01:59 (Istanbul, UTC+3).
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…