Standard OpenAI Compatible

Sealarca API Documentation.

Connect your business software, local tools (Word, LibreChat), or scripts with a single line of configuration.

Public entry point

One stable base URL for OpenAI-compatible clients, with Bearer authentication.

Paramètre	Valeur	Description
Base URL	https://sealarca.ch/v1	Secure public endpoint for all OpenAI-compatible clients.
Authorization	Bearer sk-sealarca-...	Send your Sealarca key in the Authorization header.
GET Models	GET /v1/models	List of open-source models enabled for your account.
POST Chat	POST /v1/chat/completions	Main inference route in the Vault environment.

API Key Security

Store your key in a secure environment (env variables). Never expose it in public client-side code.

Execute an encrypted request

The main format is 100% OpenAI-compatible: same SDK, new base URL.

cURLbash

curl https://sealarca.ch/v1/chat/completions \
  -H "Authorization: Bearer $SEALARCA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-5.2-vault",
    "messages": [
      { "role": "user", "content": "Analyse ce texte confidentiel..." }
    ],
    "stream": false
  }'

Pythonpython

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["SEALARCA_API_KEY"],
    base_url="https://sealarca.ch/v1",
)

response = client.chat.completions.create(
    model="glm-5.2-vault",
    messages=[{"role": "user", "content": "Analyse ce texte confidentiel..."}],
)

print(response.choices[0].message.content)

TypeScript / Nodets

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.SEALARCA_API_KEY,
  baseURL: "https://sealarca.ch/v1",
});

const response = await client.chat.completions.create({
  model: "glm-5.2-vault",
  messages: [{ role: "user", content: "Analyse ce texte confidentiel..." }],
});

console.log(response.choices[0]?.message?.content);

API error handling

Standard status codes to diagnose keys, balance, or unavailability.

Code HTTP	Cause probable	Action recommandée
401	Missing, invalid, or revoked API key.	Check the Authorization header and generate a new key in the dashboard.
400	Malformed JSON payload or unsupported parameters.	Check your request structure (messages, model).
402 / 429	Credits balance exhausted or rate limit reached.	Refill your credits balance or throttle your requests.
503	Model temporarily unavailable.	Retry in a few moments or contact support.