Setup guide

Prerequisites

Quick install

The fastest path. The wizard asks for your key, configures the clients you pick, and verifies the connection before it exits.

npx opusmax

Windows — PowerShell

If you would rather run the script directly:

Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser
irm https://opusmax.pro/setup.ps1 | iex

macOS and Linux

The shell equivalent:

curl -fsSL https://opusmax.pro/setup.sh | bash

Claude Code CLI

Point the Claude Code CLI at OpusMax and every model in the lineup becomes selectable from the same session.

Automatic — recommended

Run npx opusmax and select Claude Code CLI.

Manual configuration

Create or edit ~/.claude/settings.json:

{
  "env": {
    "ANTHROPIC_AUTH_TOKEN": "YOUR_API_KEY",
    "ANTHROPIC_BASE_URL": "https://api.opusmax.pro",
    "ANTHROPIC_MODEL": "claude-fable-5[1m]",
    "ANTHROPIC_SMALL_FAST_MODEL": "claude-haiku-4-5-20251001",
    "ANTHROPIC_DEFAULT_SONNET_MODEL": "claude-sonnet-5",
    "ANTHROPIC_DEFAULT_OPUS_MODEL": "claude-opus-4-8[1m]",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "claude-haiku-4-5-20251001",
    "CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC": "1"
  },
  "hasCompletedOnboarding": true
}

VS Code

The VS Code Claude extension reads the same configuration as the CLI.

Automatic — recommended

Run npx opusmax and select VS Code.

Manual configuration

Same file as Claude Code CLI above. Restart VS Code once you have saved it.

Cursor

Route Cursor’s AI features through OpusMax.

Automatic — recommended

Run npx opusmax and select Cursor.

API routing

Open Settings → Models → add an OpenAI-compatible model with:

• Base URL: https://api.opusmax.pro/v1
• API key: your OpusMax key
• Model: claude-sonnet-5

Windsurf

Route Windsurf’s AI provider through OpusMax.

Automatic — recommended

Run npx opusmax and select Windsurf.

API routing

Open Settings → AI Provider → set the base URL to:

https://api.opusmax.pro/v1

Cline

Configure the Cline VS Code extension to use OpusMax.

Automatic — recommended

Run npx opusmax and select Cline.

Manual configuration

Add to your VS Code settings.json:

{
  "cline.apiProvider": "anthropic",
  "cline.anthropicBaseUrl": "https://api.opusmax.pro",
  "cline.apiKey": "YOUR_API_KEY"
}

Roo Code

Configure the Roo Code VS Code extension to use OpusMax.

Automatic — recommended

Run npx opusmax and select Roo Code.

Manual configuration

Add to your VS Code settings.json:

{
  "roo-cline.apiProvider": "anthropic",
  "roo-cline.anthropicBaseUrl": "https://api.opusmax.pro",
  "roo-cline.apiKey": "YOUR_API_KEY"
}

Authentication

Every proxy endpoint accepts your key through either header:

x-api-key: YOUR_API_KEY
# or
Authorization: Bearer YOUR_API_KEY

Dashboard endpoints are separate — they use a JWT from the /auth/login endpoint.

Messages

{
  "model": "claude-sonnet-5",
  "max_tokens": 1024,
  "messages": [
    { "role": "user", "content": "Hello, Claude!" }
  ],
  "stream": false
}

{
  "id": "msg_...",
  "type": "message",
  "role": "assistant",
  "model": "claude-sonnet-5",
  "content": [
    { "type": "text", "text": "Hello! How can I help you today?" }
  ],
  "stop_reason": "end_turn",
  "usage": {
    "input_tokens": 12,
    "output_tokens": 15
  }
}

Models

{
  "data": [
    {
      "id": "claude-fable-5",
      "type": "model",
      "display_name": "Claude Fable 5",
      "created_at": "2026-06-09T00:00:00Z",
      "context_window": 1000000
    },
    {
      "id": "claude-opus-5",
      "type": "model",
      "display_name": "Claude Opus 5",
      "created_at": "2026-07-24T00:00:00Z",
      "context_window": 1000000
    },
    {
      "id": "claude-sonnet-5",
      "type": "model",
      "display_name": "Claude Sonnet 5",
      "created_at": "2026-06-10T00:00:00Z",
      "context_window": 1000000
    },
    {
      "id": "claude-opus-4-8",
      "type": "model",
      "display_name": "Claude Opus 4.8",
      "created_at": "2026-05-28T00:00:00Z",
      "context_window": 1000000
    }
  ],
  "has_more": false
}

Truncated above. The live endpoint returns all 13 models; see the full list.

Token counting

{
  "model": "claude-sonnet-5",
  "messages": [
    { "role": "user", "content": "How many tokens is this?" }
  ]
}

Key status

{
  "status": "found",
  "name": "My Key",
  "isActive": true,
  "windowTokenLimit": "5000000",
  "windowTokensUsed": "1234567",
  "windowActive": true,
  "windowResetAt": "2026-03-19T15:00:00.000Z",
  "planName": "Pro",
  "expiresAt": "2026-06-19T00:00:00.000Z",
  "totalRequests": 142,
  "last24h": {
    "requests": 28,
    "tokensIn": 45000,
    "tokensOut": 12000,
    "totalTokens": 57000,
    "avgLatencyMs": 1250
  }
}

Web search

{
  "query": "latest Node.js release 2026"
}

Image analysis

{
  "prompt": "Describe what you see in this image",
  "image_url": "https://example.com/photo.jpg"
}

Built-in tools

Both tools run server-side. There is nothing to install on your machine and no MCP server to configure.

Available models

Every model below answers on the same key and the same base URL. Pass the ID exactly as written.

Troubleshooting