Provider: NVIDIA NIM

NVIDIA’s API catalog hosts open models behind an OpenAI-compatible endpoint at https://integrate.api.nvidia.com/v1.

Get an API Key

1. Sign in at build.nvidia.com
2. Generate a key at build.nvidia.com/settings — it starts nvapi-
3. Developer accounts get free request credits; check the site for current limits

Configure VibeCody

Option 1: Encrypted store (recommended)

vibecli set-key nvidia "..."
vibecli --provider nvidia

The key is stored encrypted in ~/.vibecli/profile_settings.db, not in any plaintext file. See Secure settings storage.

Option 2: Environment variable

export NVIDIA_API_KEY="..."
vibecli --provider nvidia

Option 3: Config file (~/.vibecli/config.toml)

[nvidia]
enabled = true
model = "nvidia/nemotron-3-super-120b-a12b"
api_url = "https://integrate.api.nvidia.com/v1"   # optional, default shown

Prefer option 1 or 2 for the key itself. In VibeCoder, the key field is under Settings → Configuration → Keys, and NVIDIA NIM appears in the toolbar model dropdown once a key is set.

Model Selection

Default: nvidia/nemotron-3-super-120b-a12b

Ids are vendor/model. The list was read from the public GET https://integrate.api.nvidia.com/v1/models on 2026-10-15 (81 entries, chat models only kept). That endpoint is the authority — rerun it to see what is current:

curl -s https://integrate.api.nvidia.com/v1/models | python3 -c "import sys,json;print('\\n'.join(sorted(m['id'] for m in json.load(sys.stdin)['data'])))"

Being listed does not mean your account can call a model.

Verify Connection

vibecli --provider nvidia -c "Write a Python async web scraper"

Troubleshooting

401

The key is missing or invalid — it must be the full nvapi-… string.

Model access error

The model is listed but your account lacks access to it. Pick another from the table.