NVIDIA

Quickly set up the NVIDIA LLM service into your NLX workspace

Integrate NVIDIA

To integrate with the NVIDIA LLM service, begin by selecting Integrations from your workspace menu:

• Click Add integration > Choose NVIDIA from the Type dropdown

• Provide a name for the integration

Go to https://build.nvidia.com/explore/discover

• Select a desired generative AI model

• Choose Get API Key (may be prompted to log in)

• Enter the following into NLX:

  • API key

  • Base URL

  • Optional Model name: Enter the preferred model (.e.g, "meta/llama-3.1-405b-instruct"). Alternatively, you may specify the model while invoking the LLM in conversation using a Generative node type

• When ready, click Create Integration

At any time, you may edit or delete your NVIDIA integration by expanding the integration name and selecting the edit or delete links.

---

Using NVIDIA

To successfully use your NVIDIA integration with conversations created in your workspace, be sure to complete the following:

• create a flow

• add a Generative node type to the Canvas and assign NVIDIAI as the integration > Provide your prompt (follow instructions under Side panel tab of Generative node instructions)

• create a build & deploy