Fast inference for low latency applications
Developers expect the best performance without compromise. SambaRack delivers fast inference on the best open-source models like OpenAI’s gpt-oss and DeepSeek. This allows developers to focus on their application, while we handle the rest.
Best energy efficiency
SambaNova’s architecture drives 4x better energy efficiency over traditional GPUs, reducing physical footprint and power consumption. This helps OVHcloud meet its sustainability AI goals.
Model bundling reduces physical footprint
SambaNova’s 3-tier memory architecture enables OVHcloud to serve more models with less hardware. Multiple models can be served per SambaRack and can be hot-swapped at runtime with very low switching times.