Product
AI Video Generation API for SaaS
Connect video models with ModelGo, run concurrent jobs, and track API costs. Book an enterprise demo and apply for up to $200 in load-testing credits.
Add AI video generation to your video product. Access multiple models with one ModelGo API key, submit concurrent jobs, and track usage and costs in one place. Enterprise teams get dedicated support from model evaluation and API integration through production use.
Discuss your product and expected workload with dedicated sales. Apply for up to $200 in enterprise load-testing credits.
- Multiple video models
- One API key and centralized usage
- Usage-based billing
- Higher concurrency without a plan upgrade
- Dedicated enterprise support
- Evaluation, integration, and production issues
In this article
Build video generation into your product
Let users enter a prompt or upload a reference image, submit a generation task, and preview the returned video in your app. Your team builds the product experience. ModelGo provides the video model APIs.
Sample promptA cargo bike on a morning delivery route.
View the cURL example
curl https://api.modelgo.ai/v1/content_generation/tasks \
-H "Authorization: Bearer $MODELGO_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "doubao-seedance-2.0-mini",
"content": [{"type": "text",
"text": "A cargo bike on a morning delivery route"}],
"duration": 5, "ratio": "16:9", "resolution": "720p"
}'Set MODELGO_API_KEY in your terminal. Use the returned id to check the task status.
Connect and evaluate video models
Adding model options can mean managing more provider accounts, credentials, and usage records. ModelGo gives your team one API key and a single place to manage usage across supported video models from Seedance, Kling, MiniMax, and Wan.
Compare output quality, completion time, and cost using the same prompts or reference images. Choose a model mix for your product, using each model’s documented input formats, parameters, and capabilities.
- Video platforms and editors
- Offer text-to-video, image-to-video, and model selection inside a generator or editing platform. Let users preview different results in the same project.
- 2D animation and motion design
- Turn reference images into video clips inside animation and motion tools, supporting animatics and previews of movement and timing.
- 3D, VFX, and previsualization
- Generate camera and lighting reference videos inside previs tools or internal production systems, to guide the scenes artists build in 3D software.
Run video jobs concurrently
When users generate videos at the same time, or your product runs a batch workflow, each job needs its own progress and result. ModelGo accepts generation tasks asynchronously and returns an ID for each one, so your app can track status and deliver completed videos.
- Submit concurrent jobs
Send generation requests for your workload and save each task ID.
- Show progress
Check each task so users can see what’s queued, ready, or needs attention.
- Deliver each clip
Make completed videos available to users while other jobs continue running.
Enterprise video generation has no fixed concurrency cap and requires no plan upgrade to unlock concurrent jobs.
Throughput depends on the model and upstream capacity. Test your own inputs and expected workload to evaluate completion rates, batch completion time, and generation costs.
Enterprise load-testing credits. Work with dedicated sales and technical support to plan tests around your models, representative inputs, and expected peak workload.
Book a DemoTrack video generation costs
Generating a usable video can take several attempts. When budgeting for a video feature, account for retries and unused outputs as well as the price of a single request.
10.13 s · 1920 × 1088 · $0.753 API charge in this test.
Cost per usable clip
Settled API spend ÷ clips accepted in your product
Pay by usage. Each model has its own billing units. Review settled charges and usage by model, workspace, or API key.
Investigate failed jobs. Share the request ID with ModelGo support.
Dedicated support from evaluation to production
- Model evaluation
- Discuss your product, model shortlist, and expected usage with dedicated sales. Define the quality, timing, and cost you need to test.
- API integration
- Build with documented endpoints and request examples. Get technical support for integration issues and task-status questions.
- Production support
- Share the request ID and observed issue. Technical support can help review request details, investigate failed jobs, and troubleshoot production issues.
Case Study
How Besideo added batch video generation
Besideo integrated the ModelGo video API so users can turn pet photos into personalized clips inside its app. The team can try different models and run batches through the same integration.
- Batch generation
- In one measured batch, all 9 concurrent jobs returned videos in .
- Model access
- 6 video models used through one API integration.
Discuss your video API integration
Work with ModelGo’s dedicated sales and technical support to turn your product requirements, model options, and peak workload into a test plan.
Bring your product requirements
- The video features you want to build
- Your model shortlist and representative inputs
- Expected usage, peak job volume, and launch plans
Enterprise consultation and testing
Book a DemoBook a demo to contact dedicated sales and apply for up to $200 in load-testing credits. Our team will help assess model integration and concurrency testing needs, with technical support during integration and ongoing use.
Your engineering team can also explore the API independently.
Get $0.50 on signup, $2 more for completing starter tasks, and up to $10 on your first top-up. Request enterprise load-testing credits through your dedicated sales contact via Book a Demo.
FAQ
Do I need a plan upgrade for higher video concurrency?
No plan upgrade is required for higher video concurrency. Enterprise video generation has no fixed platform-wide concurrency cap. Actual capacity depends on the model and upstream availability. Book a demo to discuss load testing against your expected workload with dedicated sales.
How do I request an enterprise load test?
Use Book a Demo to contact ModelGo VIP Sales and request up to $200 in enterprise load-testing credits. Share your model shortlist, sample shots, and expected peak job volume so the team can help plan your test.
What support do enterprise customers get?
Dedicated sales can help discuss your product, model shortlist, expected usage, and load-testing needs. During integration and after launch, technical support can use request IDs to review request details, investigate failed jobs, and troubleshoot production issues.
How is video generation API usage billed?
Billing units and rates depend on the model and settings. For example, Wan 2.7 bills by delivered seconds and resolution tier. Check the model’s pricing in the console. For an enterprise rollout, your dedicated sales contact can discuss your model mix and expected usage.
Can I add both text-to-video and image-to-video to my app?
Yes. Use a supported model and request format for each workflow. For example, Wan 2.7 on ModelGo offers a text-to-video model for prompts and an image-to-video model that takes a first frame.
What credits can new ModelGo users get?
Get $0.50 on signup, $2 more for completing starter tasks, and up to $10 on your first top-up. The combined reward is up to $12.50.