27 public models
Compare video models
Compare generation modes, durations, formats, estimated credits, practical uses, and prompting guidance before opening a model in Starpop.
A
Alibaba
1 model
ByteDance
4 models
Seedance 1.5 ProA ByteDance video model for native-audio shots across a particularly broad set of social and cinematic aspect ratios.Seedance 2ByteDance’s full multimodal Seedance route for production that combines text, images, video references, and audio.Seedance 2 FastA faster Seedance 2 route for multimodal iteration at 480p or 720p across common and ultrawide formats.Seedance 2 MiniThe lowest-cost Seedance 2 route for mixed-input drafts and scalable video variation at 480p or 720p.
Creatify
1 model
5 models
Gemini Omni FlashA multimodal Google video model that can generate from text or images and transform an existing source video.Veo 3.1Google’s cinematic video model for polished text-to-video and image-to-video shots with native audio.Veo 3.1 (Slow)A lower-credit route to Google Veo 3.1 for teams that can exchange turnaround time for production cost.Veo 3.1 FastA faster, lower-credit Veo route for creative iteration and production shots that still need native audio.Veo 3.1 LiteThe most credit-efficient Veo 3.1 option in Starpop for drafts, testing, and scalable social production.
Kling AI
8 models
Kling 2.6 Motion ProThe higher-quality Kling 2.6 motion-transfer option for approved character performances and more polished delivery.Kling 2.6 Motion StandardA cost-conscious motion-transfer model that maps movement from a reference clip onto a character image.Kling 2.6 ProA focused Kling generation model for high-quality five- or ten-second text-to-video and image-to-video shots.Kling 3.0 Motion ProKling’s higher-quality 3.0 motion-control route for final character performances based on a reference clip.Kling 3.0 Motion StandardThe standard Kling 3.0 motion-control route for transferring a reference performance while keeping production costs measured.Kling 3.0 ProKling’s premium 3.0 generation route for cinematic text- and image-guided video with native audio.Kling 3.0 StandardA cost-efficient Kling 3.0 route for native-audio video from three to fifteen seconds.Kling O3 ProA broad Kling production model that combines generation, visual references, and source-video editing in one route.
OpenAI
2 models
Pruna AI
1 model
Sync Labs
1 model
VEED
3 models
Fabric 1.0A portrait-animation model for creating talking-head video from a still image and an audio track.Fabric 1.0 FastThe faster Fabric queue for time-sensitive talking-head production from a portrait and recorded speech.VEED SubtitlesA video finishing tool for rendering readable, styled captions directly into an existing clip.
Choosing a model
Start with the job, not the model name.
Video models solve different production problems. Some create a complete shot from text, while others animate an image, transfer motion, replace dialogue, or finish an existing video. Start with the input you already have and the format you need to deliver.
Open the Starpop workspace