GPT-6 Astra 助 AI 视频时长升至10分钟
OpenAI 发布与 AI 视频平台 Higgsfield 的合作案例，称接入 GPT-6 Astra 后，Higgsfield 生成的视频时长上限从约 1 分钟提升到 10 分钟以上。Higgsfield 联合创始人兼 CEO Alex Mashrabov 表示，Astra 让创意智能体能够调用外部工具处理 3D 建模等复杂工作，从而在长视频中保持角色、场景和视觉细节的一致性。Higgsfield 称其拥有近 50 万订阅者的 YouTube 频道已用该流程产出超过 10 分钟的教学视频，公司还集成了 GPT-Image-2.5 用于稳定渲染文字与信息图，并用 Codex 加快从创意到软件原型的过程；下一步计划基于 GPT-Live-1 探索“语音到视频”。上述时长与效果均为厂商说法，报道未提供独立测量。
这是目前少有的、具体说明 GPT-6 Astra 如何被用于解决长视频中角色与场景一致性难题的案例，但所有能力与时长数字均来自 OpenAI 与 Higgsfield 双方的自述。
Higgsfield AI(opens in a new window) is on a mission to make professional video and image creation accessible to anyone with an idea. Before GPT‑6 Astra, those videos topped out at around one minute. Now, says CEO and co-founder Alex Mashrabov, users can create videos over 10 minutes long.
Longer videos are just one way Higgsfield AI helps smaller creative teams take on more ambitious work. According to Alex, with models like GPT‑6 Astra and GPT‑Image‑2.5, it’s now possible for a single creator to execute high-end fashion campaigns from start to finish.
“What’s most important for us at Higgsfield is to see that smaller creative teams can actually be more successful now with the power of models like GPT-6 Astra and GPT-Image-2.5.”
—Alex Mashrabov, CEO and co-founder, Higgsfield AI
Making longer videos possible with GPT‑6 Astra
Higgsfield AI’s creative agents take on repetitive production tasks, such as creating multiple variations of a shot. With GPT‑6 Astra, those agents can also call external tools to handle complex, manual work in 3D modeling. Alex says that capability helps address one of creative AI’s hardest problems: maintaining consistency across longer videos.
#OpenAI#Higgsfield AI#GPT-6 Astra#GPT-Image-2.5#Codex#GPT-Live-1