Skip to content

Features

An overview of Brivionix's current capabilities based on the live platform.

OpenAI-Compatible API

All models are accessed via the standard OpenAI Chat Completions endpoint:

text
POST https://brivionix.com/v1/chat/completions

Any OpenAI-compatible SDK, client, or framework works out of the box — just change the Base URL.

Available Models

ModelContextStrengths
gpt-5.51.1MFlagship — complex reasoning, longest context
gpt-5.4400KHigh-end balanced performance
gpt-5.4-mini400KBest value — recommended for daily use
gpt-5.3-codex400KOptimized for code generation
gpt-5.2400KCost-effective general purpose

All models support: Reasoning, Tool Calling, Vision, Streaming, and Cache Billing (~90% discount on cached tokens).

Smart Routing

  • Weighted traffic distribution across upstream channels
  • Automatic retry on failure
  • Load balancing for high availability

Pay-As-You-Go Pricing

  • 0.85 RMB = 1 API Credit
  • Supports Stripe
  • Volume discounts from 3% to 15%
  • Cache billing: ~90% off for cached tokens

Token Management

  • Create, edit, disable, and delete tokens
  • Set per-token quota limits
  • Model whitelist per token
  • IP restrictions and expiration dates
  • Group-based model access

Client Integration

Cherry Studio, AionUI, OpenCat — click a button in the console to auto-configure.

Manual Setup

ChatGPT Next Web, Lobe Chat, Open WebUI, Cursor, VS Code Continue — set the Base URL and API Key.

Console Navigation

Top Navigation:

Menu ItemDescription
HomeReturn to the main website
ConsoleEnter the user console
PricingView model pricing (login required)
AboutLearn about the platform

Console Sidebar:

Menu ItemFunction
PlaygroundInteractive API testing with live results
ChatBuilt-in chat interface
DashboardBalance, usage stats, model analytics
Token ManagementCreate and manage API Keys
Usage LogsPer-request details: tokens, cost, IP
Drawing LogsImage generation history
Task LogsAsync task history
Wallet ManagementTop up, billing, redemption codes
Personal SettingsAccount info and preferences

Support