n8n-nodes-cerebras-api

This is an n8n community node that integrates Cerebras AI inference capabilities into your n8n workflows. Get ultra-fast AI inference using Cerebras' Wafer Scale Engine (WSE) technology with familiar OpenAI-compatible APIs.

🚀 Features:

• Lightning-fast inference with Cerebras' WSE technology
• OpenAI-compatible API (seamless migration from OpenAI nodes)
• Support for both chat completions and text completions
• Latest Cerebras models (GPT-OSS-120B, Llama-4, Qwen-3, etc.)
• Advanced parameter controls (temperature, top-p, penalties, streaming)
• Professional Cerebras-style branding and UI
• Comprehensive error handling and validation

Installation

Follow the installation guide in the n8n community nodes documentation.

Install from npm

npm install n8n-nodes-cerebras-api

Install in n8n

1. Go to Settings → Community Nodes in n8n
2. Click "Install"
3. Enter: n8n-nodes-cerebras-api
4. Click "Install" and wait for completion

Prerequisites

• n8n version 1.0.0 or higher
• A Cerebras account with API access
• Valid Cerebras API key

Configuration

1. Get Your Cerebras API Key

1. Sign up for a Cerebras account
2. Generate an API key from your dashboard
3. Keep your API key secure

2. Set up Credentials in n8n

1. In n8n, go to Settings → Credentials
2. Click "Create New"
3. Search for "Cerebras API"
4. Enter your API key
5. Test the connection
6. Save the credentials

Available Models

• GPT OSS 120B (gpt-oss-120b) - 65,536 max tokens
• Llama 3.3 70B (llama-3.3-70b) - 65,536 max tokens
• Llama 4 Maverick 17B (llama-4-maverick-17b-128e-instruct) - 8,192 max tokens
• Llama 4 Scout 17B (llama-4-scout-17b-16e-instruct) - 8,192 max tokens
• Llama 3.1 8B (llama3.1-8b) - 8,192 max tokens
• Qwen 3 235B Instruct (qwen-3-235b-a22b-instruct-2507) - 64,000 max tokens
• Qwen 3 235B Thinking (qwen-3-235b-a22b-thinking-2507) - 65,536 max tokens
• Qwen 3 32B (qwen-3-32b) - 65,536 max tokens
• Qwen 3 Coder 480B (qwen-3-coder-480b) - Large context model

Usage

Chat Completion

The most common use case - have conversations with AI:

{
  "operation": "chatCompletion",
  "model": "llama3.1-8b",
  "messages": [
    {
      "role": "system", 
      "content": "You are a helpful AI assistant."
    },
    {
      "role": "user",
      "content": "Explain quantum computing in simple terms."
    }
  ]
}

Text Completion

For prompt-based completions:

{
  "operation": "textCompletion",
  "model": "gpt-oss-120b",
  "prompt": "The future of artificial intelligence is"
}

Advanced Parameters

Fine-tune the AI's behavior:

• Temperature (0-2): Controls creativity/randomness (default: 0.7)
• Max Tokens (1-65536): Maximum response length (varies by model)
• Top P (0-1): Nucleus sampling parameter (default: 1.0)
• Frequency Penalty (-2 to 2): Reduce repetition (default: 0)
• Presence Penalty (-2 to 2): Encourage topic diversity (default: 0)
• Stop Sequences: Custom stop words (comma-separated)
• Stream Response: Real-time response streaming (default: false)

Example Workflows

Simple Chat Assistant

Create a workflow with:

1. Manual Trigger → Cerebras AI
2. Configure Cerebras AI node:
   • Operation: Chat Completion
   • Model: llama3.1-8b
   • Message: User input
   • Credentials: Your Cerebras API credentials

Content Generation Pipeline

graph LR
    A[Webhook] --> B[Generate Outline<br/>Cerebras AI]
    B --> C[Write Content<br/>Cerebras AI]
    C --> D[Format Output<br/>Set Node]

Code Analysis Workflow

graph LR
    A[HTTP Request<br/>Get Code] --> B[Analyze Quality<br/>Cerebras AI]
    B --> C[Generate Suggestions<br/>Cerebras AI]
    C --> D[Send Report<br/>Email]

API Response Format

The node returns the complete Cerebras API response, following OpenAI's format:

Chat Completion Response

{
  "id": "chatcmpl-xxx",
  "object": "chat.completion",
  "created": 1677652288,
  "model": "llama3.1-8b",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "The AI response content here..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 25,
    "completion_tokens": 150,
    "total_tokens": 175
  }
}

Text Completion Response

{
  "id": "cmpl-xxx",
  "object": "text_completion", 
  "created": 1677652288,
  "model": "gpt-oss-120b",
  "choices": [
    {
      "text": "The completion text here...",
      "index": 0,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 90,
    "total_tokens": 100
  }
}

Performance Benefits

Cerebras offers significant advantages over traditional cloud AI APIs:

• 🚀 Up to 16x faster inference compared to GPU-based solutions
• ⚡ Consistent low latency with no cold starts
• 📈 High throughput for batch processing
• 💰 Cost-effective for high-volume use cases
• 🔄 Reliable performance with dedicated hardware

Rate Limits

All models share these rate limits:

• Requests: 30/minute, 900/hour, 14,400/day
• Tokens: Varies by model (see Available Models section)

Error Handling

The node provides comprehensive error handling:

• 401 Unauthorized: Invalid API key
• 429 Rate Limited: Exceeded rate limits
• 400 Bad Request: Invalid parameters
• 500 Server Error: Cerebras API issues

Enable "Continue on Fail" in node settings to handle errors gracefully in workflows.

Troubleshooting

Common Issues

1. Authentication Error

   • Verify your API key is correct
   • Check that your Cerebras account has API access
   • Ensure credentials are properly saved in n8n

2. Model Not Found

   • Use exact model names from the Available Models list
   • Check Cerebras documentation for model availability

3. Request Timeout

   • Try reducing the max_tokens parameter
   • Check your network connection
   • Consider using a smaller model for testing

4. Rate Limits Exceeded

   • Implement delays between requests in workflows
   • Monitor your usage in the Cerebras dashboard
   • Consider upgrading your Cerebras plan

Getting Help

• Cerebras Documentation: inference-docs.cerebras.ai
• n8n Community Forum: community.n8n.io
• GitHub Issues: github.com/kevint-cerebras/n8n-cerebras

Development

Building from Source

git clone https://github.com/kevint-cerebras/n8n-cerebras.git
cd n8n-cerebras
npm install
npm run build

Testing Locally

npm link
n8n start
# Node will appear in n8n interface

Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

License

MIT - see LICENSE file for details.

Changelog

0.2.1

• Fixed: Package name to comply with n8n requirements (n8n-nodes-cerebras-api)
• Updated models to current Cerebras offerings
• Added GPT-OSS-120B, Llama-4, and Qwen-3 models
• Improved Cerebras-style branding and logo
• Enhanced error handling and validation

0.2.0

• Attempted clean package name (discovered n8n naming requirements)
• Professional Cerebras wafer-style logo
• All current Cerebras models supported
• OpenAI SDK integration for reliability

0.1.0

• Initial release
• Basic chat completion and text completion support
• Cerebras API integration
• n8n community node structure

---

Built for the n8n community with ⚡ Cerebras speed

Experience lightning-fast AI inference in your n8n workflows!