Welcome to the Strontium blog

strontium.ai is live, and so is this blog. Here is what Strontium does, where to start, and what you will read here.

Frontier multimodal AI, made affordable

Strontium makes image, video, audio and lipsync models available for production work. You reach them through one API and a playground, and they are ready for AI agents. We train, serve and run the stack behind them ourselves.

Where to start

• The API. Pick a model, send your inputs, and the finished output arrives on your webhook. You pay per job from a prepaid balance, with no servers to run.
• Image to video. Turn a photo into a moving clip. The product, the person and the scene stay as photographed.
• Realistic lipsync and AI avatars. Give us a voice and a face, and the mouth follows the audio. Or send a product shot and a brief, and a presenter made by our models comes back with the product in hand.
• Ad creation. Send a product URL or photos, and our team hands back finished video ads in the formats your channels take.
• Custom services. Custom inference, custom training and compute, built around your workload. We size it, build it and keep it running.

What we will write about

News about the platform and its models, and write-ups from the team that trains, serves and runs the stack: how we work, what we measure, and the evidence behind each claim. The Research page lists the areas we work on.

Talk to us

To use Strontium, request access. For custom work, talk to us. If you would like to work on this with us, see careers.