1. Install
Download the package for your distribution from repo.eldric.ai. For Fedora / RHEL:
sudo dnf install eldric-controller eldric-edge eldric-workerd eldric-datad
2. Configure
Start the four services. Each has a sane default config under /etc/eldric/. The controller listens on port 8880, the edge on 443.
3. Create the admin user
Open the chat interface in your browser (the edge serves it at /chat). On a fresh install you are prompted to create the first user. That first user becomes the system administrator — they get the full set of tenant, user, license and configuration controls. There is no shipped default password; the admin is whoever signs up first. Make sure the right person on your team signs up first, and lock down the initial signup once they are in.
4. First API call
From the admin UI, mint an API key. Then, once an inference worker is running and a model is loaded:
curl https://eldric.local/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "llama-3.3-70b",
"messages": [{"role":"user","content":"Hello."}],
"stream": true
}'
5. Use the OpenAI SDK
from openai import OpenAI
client = OpenAI(
base_url="https://eldric.local/v1",
api_key="$API_KEY",
)
resp = client.chat.completions.create(
model="llama-3.3-70b",
messages=[{"role":"user","content":"Hello."}],
)
That is it. Existing OpenAI tooling — LangChain, LlamaIndex, Continue, Cursor, whatever you have — works unchanged after pointing the base URL at Eldric.
