Skip to main content

Module nvidia

Module nvidia 

Source
Expand description

NVIDIA NIM Provider - Enterprise GPU Inference

NVIDIA NIM (NVIDIA Inference Microservices) provides optimized inference for LLMs on NVIDIA GPUs. Supports both self-hosted and cloud deployments.

Setup:

  1. Get API key from: https://build.nvidia.com
  2. Configure rco: rco config set RCO_AI_PROVIDER=nvidia RCO_API_KEY=<key> RCO_MODEL=meta/llama-3.1-8b-instruct

Docs: https://docs.nvidia.com/nim/

Structsยง

NvidiaProvider
NvidiaProviderBuilder
ProviderBuilder for NVIDIA NIM