Your own AI API. Private, fast, self-hosted.

Saduni AI serves the Saduni Lite 1.0 model behind an OpenAI-compatible endpoint. Create a key, send a request, and ship — without handing your data to anyone else.

Create an account Sign in

Self-hosted inference

Runs entirely on our own hardware. Your prompts never leave the box — no third-party model providers in the path.

OpenAI-compatible API

Drop-in /v1/chat/completions. Point any OpenAI SDK at our base URL, swap the key, and you're live.

Usage you can see

Every request is logged per key. Track tokens, latency, and volume from a dashboard built for the job.

One endpoint. Any SDK.

The API speaks the OpenAI protocol, so existing tooling works unchanged. The model always reports as saduni-lite-1.0.

curl https://aistudio.menukaabhiman.dev/v1/chat/completions \
  -H "Authorization: Bearer sk-saduni-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "saduni-lite-1.0",
    "messages": [{"role": "user", "content": "Hello"}]
  }'