Bulk Content Production
Use text, image, and video models to generate short video scripts, cover images, ad creatives, and social media content, with a single API Key handling the entire workflow, so you don’t need to move assets between multiple platforms.
Typical Workflow
- Scripts/Copywriting: Use chat models to batch-generate scripts, titles, and tags based on selected topics;
- Cover/Illustrations: Hand the key points of the copy to an image model to generate cover images and illustrations;
- Final Video: Use a video model to directly generate short video clips based on images or text.
Recommended Models
| Stage | Recommendation | Description |
|---|---|---|
| Copy generation | gpt-5.4-mini, gemini-3.5-flash | Prioritize the low-cost tier for batch tasks |
| Quality review / rewriting | claude-sonnet-5 | More stable for long-form content and style control |
| Cover/Illustrations | gpt-image-2-fast, gemini-3.1-flash-image | -fast is suitable for large-scale image generation |
| Short video | omni_flash_6s_portrait, veo_3_1_t2v_fast | Use the -portrait model directly for vertical video |
Implementation Notes
- When calling in batch, pay attention to concurrency and rate limits, and use exponential backoff retries for failed requests;
- Image/video generation is billed per request. First validate the prompt template with a small batch before scaling up; see Pricing;
- Video generation is an asynchronous task. Poll for results after submission; see Video Generation (Async Protocol).