Bring your NVIDIA API key

Put your authorized NVIDIA_API_KEY on a router you control. NVIDIA provides model access; TEKIZ.AI handles routing and request history.

OpenAI-compatible setup

For the hosted API, point OpenAI-compatible apps at https://api.tekiz.ai/v1. Hosted requests use routes available to your account. To use your own NVIDIA NIM credentials, configure them on a local router.

When NVIDIA NIM rate-limits

If an NVIDIA key or model is rate-limited, TEKIZ.AI can try another authorized route you configured.

Send a test request

This tests your hosted TEKIZ.AI key with automatic routing. To use NVIDIA NIM with your own credentials, set up a local router and configure the credentials on that router.

export OPENAI_BASE_URL=https://api.tekiz.ai/v1
export OPENAI_API_KEY=sk_tekiz_your_generated_key

curl "$OPENAI_BASE_URL/chat/completions" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "tekizai/auto",
    "messages": [{"role": "user", "content": "Explain how to size an inference server."}]
  }'

Credential load balancing across your keys

Add more than one NVIDIA API key. TEKIZ.AI spreads requests across keys that are working and pauses only a key that is exhausted or rate-limited.

Send files to services that accept them

Images, audio, video, and documents are sent only to models that can handle that kind of input.

Profiles avoid incompatible models

Automatic profiles can include NVIDIA NIM without sending images, audio, video, or documents to models that cannot handle them.

Why put TEKIZ.AI in front of NVIDIA NIM?

Your tools keep one address while TEKIZ.AI checks your keys, sends files to compatible models, and uses an available local or cloud backup when NIM cannot take the request.

Next steps

Create an API key, copy the hosted quickstart, browse NVIDIA-capable routes, or read the NVIDIA NIM integration guide.

Stay within your approved services

If your company approves only NVIDIA AI, keep that boundary. TEKIZ.AI can reduce token burn by choosing eligible models. Your NVIDIA account handles provider billing.