Free credits / metered
nvidia/NVIDIA NIM
NVIDIA NIM — free for Developer Program members.
build.nvidia.comQuick setup
Open the kRouter dashboard at http://localhost:20128
Go to Providers → click NVIDIA NIM
Paste your API key, click Save
Test the connection
Use in your IDE
Point any OpenAI-compatible client at kRouter using your local API key.
# Endpoint
http://localhost:20128/v1
# API key (from Dashboard → API Keys)
sk-krouter-XXXX
# Model
nvidia/nvidia/nemotron-4How NVIDIA NIM billing works with kRouter
NVIDIA NIM gives you free starting credits, then bills per token. kRouter tracks the remaining quota and fails over to another provider before you hit a wall. kRouter authenticates it via apikey. 4 models are routable as nvidia/<model>.
Through kRouter, NVIDIA NIM serves chat and code completion, text-to-speech, exposed on /v1/audio/speech, embeddings, exposed on /v1/embeddings. Requests reach it through the same local OpenAI-compatible endpoint as every other provider, so switching to or from NVIDIA NIM is a one-line model change in your client rather than a rewrite.
Free access for NVIDIA Developer Program members (prototyping & testing).
NVIDIA NIM models & pricing
These are the model ids kRouter routes to. This provider does not publish per-token rates.
| Model | Route as |
|---|---|
| Minimax M2.7 | nvidia/minimaxai/minimax-m2.7 |
| GLM 4.7 | nvidia/z-ai/glm4.7 |
| NV EmbedQA E5 v5 | nvidia/nvidia/nv-embedqa-e5-v5 |
| Parakeet CTC 1.1B | nvidia/nvidia/parakeet-ctc-1.1b-asr |
NVIDIA NIM links
NVIDIA NIM FAQ
Is NVIDIA NIM free through kRouter?
NVIDIA NIM is pay-per-token: you bring your own API key and are billed by NVIDIA NIM at their rates. kRouter adds nothing on top and is free and MIT-licensed.
How do I use NVIDIA NIM with Claude Code or Cursor?
Connect NVIDIA NIM in the kRouter dashboard, then point your tool at kRouter's local endpoint (http://localhost:20128/v1) with a kRouter API key. Any OpenAI-compatible client works, and models are addressed as nvidia/<model>.
Which NVIDIA NIM models can kRouter route to?
4 models, including Minimax M2.7, GLM 4.7, NV EmbedQA E5 v5. Every one is addressed as nvidia/<model> from any client.
What happens when NVIDIA NIM is rate limited?
kRouter detects the limit, parks that account until its real reset time, and fails over to your next connected account or provider — so the request still completes instead of erroring.