ProvidersNVIDIA

Model provider

NVIDIA

NIM-hosted Nemotron and open models optimized for accelerated inference.

Try in Playground Provider website

Available through one endpoint

1 NVIDIA models

API documentation →

NVIDIANemotron Ultranemotron-ultra

NVIDIA-optimized reasoning model served through NIM.

128K context$0.60 / $1.80reasoningopen-weights