Skip to content

Foundation (SDKs)

Use this page when you call LootVM from application code. IDE and CLI tools have their own guides under Client Integrations.

Setting Value
OpenAI-compatible base URL https://api.lootvm.com/v1
Anthropic-compatible host https://api.lootvm.com (SDK appends /v1/messages)
Credential LootVM-issued API key from the Console

Never put the key in the URL. Prefer environment variables or a secret manager.

Terminal window
export OPENAI_API_KEY='your-lootvm-api-key'
export OPENAI_BASE_URL='https://api.lootvm.com/v1'
from openai import OpenAI
client = OpenAI(
api_key="your-lootvm-api-key",
base_url="https://api.lootvm.com/v1",
)
resp = client.chat.completions.create(
model="your-enabled-model",
messages=[{"role": "user", "content": "Hello from LootVM"}],
)
print(resp.choices[0].message.content)
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.LOOTVM_KEY,
baseURL: "https://api.lootvm.com/v1",
});
const resp = await client.chat.completions.create({
model: "your-enabled-model",
messages: [{ role: "user", content: "Hello from LootVM" }],
});
console.log(resp.choices[0].message.content);

Use a model id from GET /v1/models for your key. See Chat Completions.

import anthropic
client = anthropic.Anthropic(
api_key="your-lootvm-api-key",
base_url="https://api.lootvm.com",
)
msg = client.messages.create(
model="your-enabled-model",
max_tokens=256,
messages=[{"role": "user", "content": "Hello from LootVM"}],
)
print(msg.content)
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
apiKey: process.env.LOOTVM_KEY,
baseURL: "https://api.lootvm.com",
});
const msg = await client.messages.create({
model: "your-enabled-model",
max_tokens: 256,
messages: [{ role: "user", content: "Hello from LootVM" }],
});
console.log(msg.content);

The Anthropic SDK posts to {base_url}/v1/messages. Do not set base_url to https://api.lootvm.com/v1 or the path doubles. See Messages (Claude) and Authentication.

Symptom Check
401 / unauthorized Wrong or revoked LootVM key; key not loaded in the process environment
404 on chat or messages Base URL missing or doubling /v1; confirm the SDK’s expected root
Model not found Model id not in /v1/models for this key, or not enabled in the Console
Upstream-looking errors You sent an upstream vendor key to api.lootvm.com instead of a LootVM key