> ## Documentation Index
> Fetch the complete documentation index at: https://docs.portkey.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Cerebras Code MCP server

> The **Cerebras Code MCP Server (v1.3.3)** integrates high-speed code generation and editing capabilities into AI-assisted IDEs such as Claude Code, Cline, and Cursor. It allows you to use your preferred AI (Claude, Qwen, etc.) for planning and strategy, and delegate the actual code-writing and modification tasks to Cerebras' optimized models — maximizing speed while avoiding API rate limits.

This MCP server runs on the **Qwen 3 Coder** model and offers near-instant code generation with deep contextual awareness. It can be embedded within any MCP-compatible IDE for a seamless code planning-to-execution workflow.

## When you should use this server

* **Accelerate code generation** and editing within AI IDEs like Claude Code, Cursor, or Cline.
* **Offload large code tasks** from other AI assistants to Cerebras' infrastructure to avoid throttling or rate limits.
* **Enable collaborative AI workflows**, where one AI (e.g., Claude) handles reasoning and another (Cerebras) executes precise code edits.
* Generate or refactor code using **multi-file context awareness** and **visual Git-style diffs**.

Ideal for developers who work with agentic coding environments and want scalable, low-latency code operations.

## Endpoints

**1. Installation**

`npm install -g cerebras-code-mcp-server`

**2. API Key Setup**
Visit cloud.cerebras.ai to generate an API key.

**3. Run the Setup Wizard**
`cerebras-mcp --config`

**IDE Support**

* Claude Code ✅
* Cline ✅
* Cursor (Beta) ⚙️
* VS Code (Copilot integration) ⚙️

## Tools provided

### **write**

The `write` tool powers Cerebras' MCP integration and supports advanced code operations.

## Notes

* The Cerebras MCP server focuses on **speed, context, and stability** — handling bulk or complex edits efficiently without hitting OpenAI or Anthropic rate limits.
* Combines perfectly with Claude Code or Cline for **AI pair programming** — Claude plans, Cerebras executes.
* Designed for scalable multi-agent development environments.