AI 视频
AI 视频使用 ai_video。创建前读取 generation_models[]，不要按模型创建不同 endpoint 或 type。
Gemini Omni Flash
{ "type": "ai_video", "prompt": "镜头缓慢穿过雨夜未来城市，霓虹倒映在街道上，保留自然环境声", "aspect_ratio": "16:9", "video_generation": { "model": "gemini-omni-flash", "task": "text_to_video", "resolution": "720p", "duration_seconds": 8, "native_audio": true }}Omni 支持 iterate 多轮编辑，调用方不需要管理 interaction ID。
Veo 3.1 Fast
{ "type": "ai_video", "prompt": "一架纸飞机穿过明亮的现代图书馆，电影感运镜", "aspect_ratio": "16:9", "video_generation": { "model": "veo-3.1-fast", "task": "text_to_video", "resolution": "720p", "duration_seconds": 8, "native_audio": true }}Veo 3.1 Standard
{ "type": "ai_video", "prompt": "日出时分的海岸科研站，缓慢推进镜头，真实环境声", "aspect_ratio": "16:9", "video_generation": { "model": "veo-3.1-standard", "task": "text_to_video", "resolution": "720p", "duration_seconds": 8, "native_audio": true }}Veo 不支持多轮编辑。当前 Public OpenAPI 仅开放文本生成，首帧、尾帧和附件暂不接受；以 Catalog 返回能力为准。