Third-Party Clients

Brivionix is 100% compatible with the OpenAI API. Set the API Base URL to https://brivionix.com to use Brivionix with virtually any OpenAI-compatible client.

Universal Configuration

Setting	Value
API Base URL	https://brivionix.com
API Key	Your token from the console (sk-xxx...)

TIP

Some clients auto-append /v1, others require you to add it manually. If one doesn't work, try the other.

---

One-Click Clients

Download and install the app first, then go to Token Management in the console and click the client icon to auto-configure via Deep Link — no manual copy-paste needed.

Client	Platform	Download
Cherry Studio	Desktop	cherry-ai.com
AionUI	Desktop	aionui.com
CC Switch	Desktop	GitHub Releases
AI as Workspace	Web (no install)	aiaw.app
OpenCat	Desktop + iOS	opencat.app

After installing, click the corresponding icon in Token Management to write your API URL and token automatically.

Cherry Studio — Manual Setup

If the Deep Link doesn't work: Settings → Add Custom Provider → API URL: https://brivionix.com → Click "Fetch Models"

---

Grok CLI Configuration (grok-4.6 / Responses)

Grok CLI uses an independent endpoint configuration. The example below routes model requests to Brivionix's unified API endpoint.

[endpoints].models_base_url is used for model listing and primary inference requests. For Brivionix, include the /v1 suffix: https://brivionix.com/v1.

The config file is ~/.grok/config.toml on Linux / macOS and %USERPROFILE%\.grok\config.toml on Windows. Set api_key to a token created under Token Management in the Brivionix console; no extra environment variable is required.

Install or update (Linux / macOS)

curl -fsSL https://x.ai/cli/install.sh | bash
grok update
grok --version

Write ~/.grok/config.toml

Replace sk-your Brivionix API key with a token created in the Brivionix console. Never commit a real key to a repository or include it in a public screenshot.

[models]
default = "grok-4.6"
default_reasoning_effort = "xhigh"

[endpoints]
models_base_url = "https://brivionix.com/v1"

[model."grok-4.6"]
model = "grok-4.6"
name = "Grok 4.6"
api_key = "sk-your Brivionix API key"
api_backend = "responses"
context_window = 500000
supports_backend_search = true

[auth]
preferred_method = "api_key"

[session]
# Optional: the official default is 85; 80 triggers auto-compaction earlier
auto_compact_threshold_percent = 80

[ui]
max_thoughts_width = 120
yolo = true
compact_mode = false
permission_mode = "always-approve"

yolo = true and permission_mode = "always-approve" let Grok CLI edit files and run commands without asking each time. Use them only in trusted projects and environments. max_thoughts_width = 120 and disabled compact_mode match the recommended local setup.

Verify the configuration and request

grok inspect
grok models
grok --model grok-4.6 --output-format json -p 'Output only XAI_GROK_OK'

context_window = 500000 is the currently published value for the official grok-4.6 model page. supports_backend_search = true is intended for the Brivionix path that forwards requests to Grok Cloud.

The api_key must be a Brivionix console token (usually prefixed with sk-), not a grok.com OAuth refresh_token. Because the key is stored in plain text, Linux / macOS users should run:

chmod 600 ~/.grok/config.toml

On shared machines, prefer Grok CLI's supported env_key method and rotate Brivionix tokens regularly.

---

Chat Clients

ChatGPT Next Web (NextChat)

Settings → Model Provider → OpenAI → Base URL: https://brivionix.com → API Key: your token → Model: gpt-5.4-mini

Lobe Chat

Settings → Language Model → OpenAI → Proxy URL: https://brivionix.com/v1 → API Key: your token

WARNING

Lobe Chat requires the /v1 suffix.

Open WebUI

Admin → Settings → Connections → OpenAI API → Base URL: https://brivionix.com/v1 → API Key: your token

---

Coding Tools

Cursor

Settings → Search "OpenAI" → Override OpenAI Base URL: https://brivionix.com/v1 → API Key: your token

TIP

Recommended models for Cursor: gpt-5.3-codex (code-optimized) or gpt-5.4-mini (all-purpose).

VS Code Continue

Edit ~/.continue/config.json:

{
  "models": [
    {
      "title": "GPT-5.3-Codex via Brivionix",
      "provider": "openai",
      "model": "gpt-5.3-codex",
      "apiBase": "https://brivionix.com/v1",
      "apiKey": "sk-your-api-key"
    }
  ]
}

---

SDK Integration

Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-api-key",
    base_url="https://brivionix.com/v1",
)

response = client.chat.completions.create(
    model="gpt-5.4-mini",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

Node.js

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-your-api-key",
  baseURL: "https://brivionix.com/v1",
});

const response = await client.chat.completions.create({
  model: "gpt-5.4-mini",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

cURL

curl https://brivionix.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-your-api-key" \
  -d '{"model": "gpt-5.4-mini", "messages": [{"role": "user", "content": "Hello!"}]}'

---

Other Compatible Clients

Any client that supports custom OpenAI API endpoints will work, including:

• BotGem (AMA) — macOS/iOS
• Chatbox — Cross-platform desktop
• LibreChat — Open-source web
• Immersive Translate — Browser translation plugin
• Bob Translate — macOS translation tool
• Raycast AI — macOS productivity

Support

Having trouble connecting? Reach out:

• 💬 Discord: discord.gg/vCSAJ9ZAa
• 📧 Email: support@brivionix.com