Quick start

Live
Create a key, discover models, and make a verified chat request.
Open in ChatGPT
(opens in new tab)
Last verified: 2026-08-11

1. Create an API key

Open Dashboard → API keys, create a key, and save it once. Store it in a server-side environment variable.
bash
export MODELRUSH_API_KEY="your-key"

2. Discover available models

bash
curl https://api.modelrush.ai/v1/models \
-H "Authorization: Bearer $MODELRUSH_API_KEY"
The response includes model IDs, modalities, limits, available regions, and a link to region discovery.

3. Send a request

bash
curl https://api.modelrush.ai/v1/chat/completions \
-H "Authorization: Bearer $MODELRUSH_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "modelrush/grok-4.6",
"messages": [{"role": "user", "content": "Return exactly: ModelRush connected"}]
}'

4. Verify the response

Keep id, model, and usage from the response. Use the request ID in Dashboard → Requests when investigating latency, errors, or cost.

Next steps

Read Chat Completions for the complete contract, or move to image, video, and audio workflows.