n8n-nodes-demeterics

Note: This is the full version for self-hosted n8n. It includes all nodes including the Demeterics Chat Model (LangChain-based). For n8n Cloud, use n8n-nodes-demeterics-lite.

An n8n community node for the Demeterics LLM Gateway - access multiple AI providers through a unified API with built-in observability, analytics, and cost tracking.

What is Demeterics?

Demeterics is an LLM observability platform that provides:

• Unified API: Access OpenAI, Anthropic, Groq, Google Gemini, and more through a single endpoint
• Cost Tracking: Monitor spend across all AI providers in real-time
• Full Observability: Log every prompt, response, and token for debugging and compliance
• Analytics Dashboard: Understand usage patterns, latency, and performance metrics

Learn more at demeterics.ai or read the documentation.

Features

• Multi-Provider Access: Switch between Groq, OpenAI, Anthropic, and Google Gemini with a single credential
• LangChain Compatible: Works seamlessly with n8n's AI Agent, LLM Chain, and other AI nodes
• Built-in Tracking: Every request is automatically logged to your Demeterics dashboard
• Cost Visibility: See exactly what each workflow costs in real-time

Installation

There are multiple ways to install this node depending on your n8n setup.

Option 1: GUI Install (Recommended for n8n Cloud & Desktop)

1. Go to Settings > Community Nodes in your n8n instance
2. Select Install
3. Enter n8n-nodes-demeterics and confirm
4. Restart n8n if prompted

For more details, see the n8n community nodes installation guide.

Option 2: Self-Hosted n8n (Docker with Custom Dockerfile)

This is the recommended approach for Docker-based self-hosted n8n servers. It bakes the node directly into your Docker image.

Step 1: Create a Custom Dockerfile

Create a Dockerfile in your n8n project directory:

FROM n8nio/n8n:latest

USER root

# Install the Demeterics node globally
RUN npm install -g n8n-nodes-demeterics

USER node

Step 2: Update docker-compose.yml

Modify your docker-compose.yml to build from the Dockerfile instead of using the pre-built image:

version: '3.8'

services:
  n8n:
    build:
      context: .
      dockerfile: Dockerfile
    # Remove or comment out the 'image:' line if present
    # image: n8nio/n8n:latest
    restart: unless-stopped
    ports:
      - "5678:5678"
    environment:
      - N8N_HOST=localhost
      - N8N_PORT=5678
      - N8N_PROTOCOL=http
      - GENERIC_TIMEZONE=America/New_York
    volumes:
      - n8n_data:/home/node/.n8n

volumes:
  n8n_data:

Step 3: Build and Start

# Build the custom image
docker-compose build

# Start n8n
docker-compose up -d

The Demeterics node will now appear in your n8n node panel.

Option 3: Self-Hosted n8n (Volume Mount Method)

If you prefer not to rebuild the Docker image, you can mount the node directly into the container.

Step 1: Install the Node Locally

# Create a directory for custom nodes
mkdir -p ~/n8n-custom-nodes
cd ~/n8n-custom-nodes

# Install the package
npm install n8n-nodes-demeterics

Step 2: Update docker-compose.yml

Add a volume mount to your docker-compose.yml:

version: '3.8'

services:
  n8n:
    image: n8nio/n8n:latest
    restart: unless-stopped
    ports:
      - "5678:5678"
    volumes:
      - n8n_data:/home/node/.n8n
      # Mount the custom node
      - ~/n8n-custom-nodes/node_modules/n8n-nodes-demeterics:/home/node/.n8n/custom/node_modules/n8n-nodes-demeterics

volumes:
  n8n_data:

Step 3: Restart n8n

docker-compose restart

Option 4: Self-Hosted n8n (Manual npm Install - Non-Docker)

For bare-metal or VM installations running n8n directly with Node.js:

Step 1: Navigate to n8n Directory

cd ~/.n8n

Step 2: Install the Package

npm install n8n-nodes-demeterics

Step 3: Restart n8n

# If running as a service
sudo systemctl restart n8n

# Or if running manually, stop and start again
n8n start

Option 5: Install from GitHub (Development/Testing)

To install directly from the GitHub repository:

# Clone the repository
git clone https://github.com/bluefermion/n8n-nodes-demeterics.git
cd n8n-nodes-demeterics

# Install dependencies
pnpm install

# Build the package
pnpm build

# Create a tarball
npm pack

# Install in your n8n custom directory
cd ~/.n8n
npm install /path/to/n8n-nodes-demeterics-0.1.0.tgz

Or install directly from GitHub:

cd ~/.n8n
npm install git+https://github.com/bluefermion/n8n-nodes-demeterics.git

Verifying Installation

After installation, verify the node is available:

1. Open your n8n instance
2. Create a new workflow
3. Click the + button to add a node
4. Search for "Demeterics"
5. You should see Demeterics Chat Model in the results

If the node doesn't appear, try:

• Restarting n8n completely
• Checking the n8n logs for errors: docker-compose logs n8n
• Ensuring the package is installed in the correct location

Getting Started

1. Get Your API Key

1. Sign up at demeterics.ai
2. Navigate to API Keys in your dashboard
3. Create a new API key
4. Note both the prefix (e.g., dmt_abc123) and secret parts

2. Add Credentials in n8n

1. Go to Credentials in n8n
2. Click Add Credential
3. Search for "Demeterics API"
4. Enter your single Demeterics API key (e.g., dmt_xxx...)
5. Optional (BYOK): Toggle BYOK and add the provider key(s) you plan to use:
   • Groq, OpenAI, Anthropic, Gemini, OpenRouter The Chat node will automatically combine keys depending on the selected provider.

3. Use in Your Workflow

The Demeterics Chat Model node works with:

• AI Agent - Build intelligent agents with tool use
• Basic LLM Chain - Simple prompt → response workflows
• Summarization Chain - Summarize documents
• Question and Answer Chain - RAG-based Q&A

Supported Providers & Models

| Provider | Models | |----------|--------| | Groq | Llama 3.3 70B, Llama 4 Maverick/Scout, Compound AI, Qwen3 32B, Kimi K2, GPT-OSS | | OpenAI | GPT-5/Mini/Nano/Codex, GPT-4.1/Mini/Nano, GPT-4o/Mini | | Anthropic | Claude Opus 4.5/4.1, Claude Sonnet 4.5/4/3.7, Claude Haiku 4.5/3.5 | | Google | Gemini 3 Pro, Gemini 2.5 Pro/Flash, Gemini 2.0 Flash, Gemini 1.5 Pro/Flash | | OpenRouter | openrouter/auto, plus curated Anthropic/Google/Meta/Qwen/Mistral routes |

See the full model list for all supported models and pricing.

Configuration Options

| Option | Description | Default | |--------|-------------|---------| | Temperature | Controls randomness (0-2) | 0.7 | | Max Tokens | Maximum response length | 4096 | | Top P | Nucleus sampling parameter | 1 | | Frequency Penalty | Reduce repetition (-2 to 2) | 0 | | Presence Penalty | Encourage new topics (-2 to 2) | 0 | | Timeout | Request timeout in seconds | 60 |

Example Workflows

[Chat Trigger] → [AI Agent] ← [Demeterics Chat Model]
                     ↑
                  [Tools]

1. Add a Chat Trigger node
2. Add an AI Agent node
3. Connect Demeterics Chat Model to the AI Agent's "Chat Model" input
4. Select your provider and model
5. Add tools as needed (Calculator, Code, HTTP Request, etc.)

Submit and Retrieve Conversion Outcomes

[Any Flow] → [Demeterics Conversion]

• Operation: Submit Outcome or Get Outcome
• Fields: cohortId (required), optional outcome, outcomeV2, label, eventDate

Export Interaction Data

[Any Flow] → [Demeterics Extract]

• Operation: Export Interactions (Simple), Create Export Job, or Stream Export by Request ID
• Formats: JSON (parsed items), CSV/Avro (binary file)

Resources

• Demeterics Platform
• Demeterics Documentation
• n8n Integration Guide
• API Reference
• n8n Community Nodes

Compatibility

• n8n: 1.0.0+
• Node.js: 18.0+

Development

# Install dependencies
pnpm install

# Build
pnpm build

# Development mode
pnpm dev

# Lint
pnpm lint

# Run n8n package scanner
pnpm check

Pack

• Install CLI (optional): make cli-install

• Build: make build

• Prepare for submission (recommended): make prepare

• Scan only: make check

• Pack tarball: make pack

• Bump version:

  • Patch: npm version patch --no-git-tag-version
  • Or set manually in package.json under version.

• Build + sanity check:

  • make build (uses n8n-node if present, falls back to tsc+gulp)
  • Optional: npm pack to inspect the tarball contents

• Publish:

  • Using pnpm script: pnpm run release
  • This runs pnpm build and pnpm publish --access public.

• Or directly with npm: npm publish --access public

  • If prompted, enter your OTP (2FA).

• Verify on npm:

  • npm view n8n-nodes-demeterics version shows the new version.

Support

• Documentation: demeterics.ai/docs
• GitHub Issues: Report a bug
• Email: [email protected]

License

MIT

Built by Demeterics - LLM Observability & Cost Tracking