Skip to main content

NVIDIA Build

NVIDIA Build hosts NVIDIA NIM endpoints for a wide catalog of open and NVIDIA-optimized models through an OpenAI-compatible API.

Getting started with NVIDIA Build​

  1. Sign in to build.nvidia.com.
  2. Generate an API key from your account.
  3. Pick a model from build.nvidia.com/models (for example meta/llama-3.1-70b-instruct).

Setting up NVIDIA Build in Unstract​

  • Sign in to the Unstract Platform.
  • From the side navigation menu, choose Settings 🞂 LLMs.
  • Click on the New LLM Profile button.
  • From the list of LLMs, choose NVIDIA Build.
  • Name — a unique name for this adapter instance.
  • API Key — your NVIDIA API key from build.nvidia.com.
  • Model — the model name from build.nvidia.com/models.
  • API Base — pre-filled with https://integrate.api.nvidia.com/v1; change only if NVIDIA moves the base URL.
  • Maximum Output Tokens — defaults to 4096. Leave empty to use the provider default.
  • Enable Reasoning — toggle on for reasoning models. Reveals a Reasoning Effort field (low/medium/high).
  • Leave Max Retries and Timeout at their defaults.
  • Click on Test Connection and ensure it succeeds, then click Submit.