Five minutes to first token

# Talk to AgentBox over the OpenAI-compatible API
curl http://agentbox.local:11434/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "llama3.1:8b-instruct-q4_K_M",
    "messages": [{ "role": "user", "content": "Summarize this contract." }]
  }'

from openai import OpenAI

client = OpenAI(base_url="http://agentbox.local:11434/v1", api_key="local")

resp = client.chat.completions.create(
    model="llama3.1:8b-instruct-q4_K_M",
    messages=[{"role": "user", "content": "Draft a reply."}],
)
print(resp.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "http://agentbox.local:11434/v1", apiKey: "local" });

const res = await client.chat.completions.create({
  model: "llama3.1:8b-instruct-q4_K_M",
  messages: [{ role: "user", content: "Hello, local model." }],
});
console.log(res.choices[0].message.content);

# SSH in, pull a model, run it — that's the whole setup
ssh admin@agentbox.local
ollama pull llama3.1:8b-instruct-q4_K_M
ollama run  llama3.1:8b-instruct-q4_K_M "Why local AI?"