Drop-in. One line.

Generate a key in the dashboard, point your client at api.grokified.com/v1, and send a request.

curl https://api.grokified.com/v1/chat/completions \
  -H "Authorization: Bearer $GROKIFIED_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-4.3",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Authenticate with a bearer token in the Authorization header. Keys are issued per workspace and prefixed gk_live_ or gk_test_. Never ship a key in client-side code.

Authorization: Bearer gk_live_xxxxxxxxxxxxxxxx

The API is OpenAI-compatible. Request and response shapes match the upstream xAI API exactly, including tools, JSON mode, vision inputs, and streaming.

# swap this
https://api.x.ai/v1
# for this
https://api.grokified.com/v1

POST /v1/chat/completions. The primary endpoint. Pass a model, a messages array, and optional parameters like temperature, max_tokens, tools, and response_format.

Set stream: true to receive tokens as server-sent events. The stream format is identical to upstream, so existing SSE parsers work unchanged.

Pass any of these as model. See pricing for per-token rates. The greyed slugs were retired on May 15, 2026 and now auto-redirect to grok-4.3 (or grok-build-0.1 for code), so existing code keeps working.

Beyond chat, the proxy also covers images (grok-imagine-image, grok-imagine-image-quality via /v1/images/generations), video (grok-imagine-video, grok-imagine-video-1.5 via /v1/videos/generations), and the Voice Agent, Text-to-Speech, and Speech-to-Text APIs, all at 50% of xAI list. See pricing for media and voice rates.

You’re billed at 50% of xAI list price until you’ve saved $5,000, your first $10,000 of usage at half price. After the cap, 3% off for life. Track your cap in the dashboard.

Limits scale with your xAI account tier (set by cumulative spend) and are returned on every response. Proxy overhead is under 5ms.

Errors use standard HTTP status codes with a JSON body matching the upstream schema.

Any OpenAI-compatible SDK works, just set the base URL. That includes the official openai Python and Node libraries, the Vercel AI SDK, LangChain, LlamaIndex, and most agent frameworks.