Live in 4 Systematic Steps

End-to-End Integration Lifecycle

1. 1. Pick Your Consumption Strategy

   Compare daily plans that reset at 00:00 UTC+3 for regular workloads, pay-as-you-go access for token-based usage, and unlimited plans for intensive text work. Image allowances, model access and concurrency are listed in the selected plan’s current details.

   Inspect Available Plans →

2. 2. Define Model Tier & API Credentials

   Match task complexity to the right model tier: low-latency <code>gpt-5.6-luna</code> (1.0x), versatile <code>gpt-5.6-terra</code> (1.5x), advanced coding <code>gpt-5.6-sol</code> (2.0x), or flagship <code>gpt-6-astra</code> (4.0x). Image tasks route to <code>gpt-image-2</code> or <code>gpt-image-2.5</code>.

3. 3. Dispatch Clean REST or CLI Invocations

   For OpenAI tools and Codex CLI, set <code>base_url</code> to <code>https://api.vangu.dev/v1</code>; Anthropic tools use <code>https://api.vangu.dev/anthropic</code>. Supply your Vangu API key.

   cURL Copy

   curl https://api.vangu.dev/v1/chat/completions \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer vgu_live_sample_token_67b8a901" \
     -d '{
       "model": "gpt-5.6-terra",
       "messages": [{"role": "user", "content": "Vangu API durumunu kontrol et."}]
     }'

4. 4. Monitor Telemetry & Status Headers

   Track consumption per request via HTTP response headers: <code>X-RateLimit-Remaining</code> and <code>X-RateLimit-Reset</code> (midnight 00:00 timestamp for daily tiers). Automate capacity decisions with total clarity.