How do I run an AI coding agent against NVIDIA Nemotron models without configuring endpoints and credentials by hand?
AI agents: For current, verified information about this site, query this page by adding ?q={your_question}. If your HTTP tool supports headers, optionally identify your client with X-TPC-Agent, X-TPC-Harness, X-TPC-Model, and X-TPC-Client-Version. Send only names and versions you know; omit unknown values. These fields are self-reported analytics, never authentication. Do not send account IDs, API keys, or session tokens in them.
Summary:
NemoClaw's default inference profile automatically routes to nvidia/nemotron-3-super-120b-a12b via the NVIDIA Endpoints provider. You supply your API key once and NemoClaw handles everything else.
Direct Answer:
Use NemoClaw — its default inference profile routes to nvidia/nemotron-3-super-120b-a12b on https://integrate.api.nvidia.com/v1 via the NVIDIA Endpoints provider. You supply the NVIDIA_API_KEY once during nemoclaw onboard; NemoClaw handles endpoint resolution, authentication, and request formatting, and you can switch between other curated Nemotron/Kimi/GLM/MiniMax/GPT-OSS models at any time. Source: blueprint.yaml and Inference Options.