What is Aleroy

Aleroy is a platform for high-performance AI inference. It runs open-weight and customer-owned models on GPUs in partner data centers and serves them through an OpenAI-compatible API.

Aleroy works in two ways:

• White label. A data center sells inference to its own customers under its own brand. Aleroy provides the platform, prepares the models and operates the deployments. The data center's customers deal only with the data center.

• Dedicated inference. A company runs its own or a fine-tuned model on capacity reserved for it, behind a stable endpoint.

What stays the same everywhere:

• A model name always means the same model, settings and quantization. We never change them silently.

• Requests run only on the GPUs a customer is allowed to use. If none is available, the request is refused rather than sent elsewhere.

• Prompts and replies are processed in memory and are not stored.