"vendor": "alibaba""model_name": "wan2.5-t2v-preview"Text-to-video generation with flagship quality and control
Waiting for Input
Wan 2.5 Text-To-Video Preview is Alibaba's high-quality text-to-video generation model from the Wan 2.5 series. It generates coherent videos from text prompts with support for multiple resolutions, extended durations, and automatic audio generation.
seedintegerRandom seed [0, 2147483647].
sizeEnum · Default: "1280*720"Output resolution ("width*height"). Supported tiers:
audioboolean · Default: trueEnable automatic audio generation. Set to false to force a silent output.
prompt* requiredstringText description for the desired video content (max 2000 characters).
durationEnumVideo duration in seconds (24 fps). Supported values 5 or 10.
audio_urlstringCustom audio file URL (wav/mp3, 3-30s, ≤15MB). Overrides the audio flag.
prompt_extendboolean · Default: trueEnable intelligent prompt rewriting (slightly longer latency, better detail).
safety_filterboolean · Default: trueEnable content safety filter. Defaults to true. Set to false to disable content safety inspection.
negative_promptstringNegative prompt describing unwanted content (max 500 characters).