apiex选择模型后可复制配置给 AI，也可以快速测试当前连接延迟。
Alibaba
Alibaba Tongyi Wanxiang Wan 2.6 reference-to-video model replicates the actions, effects, and camera movement of a reference video to produce up to 10 second 720p/1080p clips with synchronized audio.
Alibaba
Speed-optimized Wan 2.6 reference-to-video variant with optional audio output, replicating reference-video motion into up to 10 second 720p/1080p clips for fast, cost-effective production.
bytedance
Standard tier of the Seedance family. Supports text-to-video, image-to-video, and reference-to-video, with strong character, style, and camera consistency. Built for ads and short-form creative.
bytedance
Lightweight Seedance 2.0 variant tuned for speed and cost rather than peak fidelity, with text-to-video, first/last frame control, and multi-reference inputs.
bytedance
A video generation model built for long-form storytelling, with first-frame and first-and-last-frame control, up to 50 multimodal reference assets, and optional audio generation. It also supports video editing and extension.
kling
Fully multimodal variant of the 3.0 line, accepting text, image, and video references in a single generation, with voice-driven characters and native audio.
kling
Cinematic-oriented model with first- and last-frame control and video reference input, including edits driven by an existing clip. Suited for narrative shot sequences.
minimax
A lightweight open-weight video generation model focused on instruction-guided editing, text and brand rendering, and video-to-video motion transfer, with native audiovisual output. Built for advertising, e-commerce, and interface design workflows.