Inference

Inference services are in early access

Inference services are under development. We are seeking additional novel use cases, to help shape the service design. Please contact Stefano Schuppli at stefano.schuppli@cscs.ch and Pablo Fernandez at pablo.fernandez@cscs.ch if you are interested.

Inference services leverage Machine Learning models to generate content from user-provided inputs. We enable the deployment and operation of models and related services without requiring users to manage the underlying infrastructure.

• Managed Model Access

  Managed, Internet-accessible OpenAI/Anthropic-compatible inference APIs providing open-weight models (e.g., Apertus), with token-based resource consumption.

  LLM Inference API service