Refids
Save resources and time when deploying AI models.
Faster inference, lower costs, better for the environment.
Everything you need for deployment
The container we ship is ready-to-go for deployment and its specifications can be defined in our web app GUI – no expert knowledge required.
Model Serving
Our container comes with a RESTful HTTP prediction server, so you can make inference requests out-of-the-box to our API.
Inference Task Batching
Inference Tasks waiting to be processed are automatically batched together to increase the throughput of the model.
Model Improvements
We improve the model for deployment, resulting in faster inference and lower resource consumption.
Advanced Scheduling
Assign priorities to inference tasks and control the order in which they are processed.
CI / CD Pipeline ready
After specifying the requirements, creation and deployment of the container can be integrated into your CI / CD pipeline.
Ready for Production
Just deploy the docker container – no additional steps needed.
Get insights into where you can make improvements
- Benchmarks performed with your model
- Executed in production environment
- Valuable insights for improving your deployment pipeline and infrastructure
Get early access!
Our service is not yet publicly available. Enter your email here to get early access as soon as possible.
We won’t send you spam. Unsubscribe any time.