NVIDIA NIM

Use hosted models from NVIDIA’s API catalog.

Overview

build.nvidia.com hosts many models you can call with one key. It is meant for trying and prototyping.

1
Go to build.nvidia.com and sign in or create an NVIDIA Developer account (email verification only).
2
Open Settings → API Keys and choose Generate Key. You can also open any model and choose Get API Key.
3
Copy the key (it starts with nvapi-). It is shown once.

Cost. Free trial access is rate limited, and the amount and rules have changed over time. Check your account page for your current limits.

1
Open Loom, then Providers in the menu, and choose Add provider.
2
Pick NVIDIA NIM (the search box finds it).
3
Fill in the fields in the table below.
4
Choose Connect. Loom checks the key and says how many models it found. If it cannot check, it says so, and you can Save anyway and try a message.
5
Open the model chip below the message box and pick a model.

Field in Loom	What to put	Needed
API key	Paste the key you copied. Create a key (starts with nvapi-) at build.nvidia.com.	Yes
Endpoint	`https://integrate.api.nvidia.com/v1`	Pre-filled
Name	Anything you like. It is shown in the model picker	Optional

Appears in Loom as	NVIDIA NIM
Sign-in	Your key, sent as `Authorization: Bearer`
Models	Loom fetches the list from your account
Tools (connectors, search)	Depends on the model
Photos	Depends on the model
PDFs	No

Loom sets a sensible output limit for NVIDIA, which otherwise cuts replies short.

What you see	What to do
429	You hit a trial rate limit. Wait a minute, or try a lighter model.
404 for a model	Model names look like `meta/llama-3.3-70b-instruct`. Copy one from the catalog.