n8n-nodes-github-copilot-models

An n8n community node that lets you use GitHub Copilot chat models and GitHub Models embeddings in n8n AI workflows.

Note: This is a vibe-coded fork of kk17/n8n-nodes-github-copilot-chat-models by @kk17. The original package provides the chat model node; this fork adds an embeddings node powered by the official GitHub Models API.

Features

• Chat Model — Connect n8n AI Agent nodes to GitHub Copilot models via api.githubcopilot.com
• Embeddings — Generate vector embeddings using the official GitHub Models API (text-embedding-3-small, text-embedding-3-large, etc.)
• Dynamically fetch available models from the API (both chat and embedding)
• Support for both github.com and GitHub Enterprise (chat model)
• Single shared credential for both nodes

Prerequisites

• An active GitHub Copilot subscription
• n8n instance (self-hosted or cloud)

Installation

Via n8n Community Nodes (recommended)

1. In n8n, go to Settings > Community Nodes
2. Click Install
3. Enter n8n-nodes-github-copilot-models
4. Click Install

Manual / Development

See docs/DEVELOPMENT.md

Setup

1. Create a Credential

See docs/CREDENTIALS.md

The same GitHub Copilot API credential is used by both the chat model and embeddings nodes. The OAuth token obtained through the device flow works with both APIs.

2. Chat Model Node

1. Open an AI Agent node
2. Click the Language Model input
3. Search for GitHub Copilot Chat Model
4. Select your credential
5. Choose a model from the Model dropdown (loaded dynamically)
6. Optionally configure Temperature and Maximum Tokens

| Option | Description | Default | |--------|-------------|---------| | Model | GitHub Copilot model to use (loaded dynamically) | — | | Temperature | Controls response randomness (0–2) | 0.7 | | Maximum Tokens | Max tokens to generate. -1 = no limit | -1 |

3. Embeddings Node

1. In any node that accepts an Embedding input (e.g. vector stores, retrievers)
2. Search for GitHub Copilot Embeddings
3. Select your credential
4. Choose a model from the Model dropdown (loaded dynamically from the GitHub Models catalog)

| Option | Description | Default | |--------|-------------|---------| | Model | Embedding model to use (loaded dynamically) | — | | Dimensions | Output embedding dimensions (256–3072). Only for text-embedding-3 models. | 1536 | | Batch Size | Max documents per API request | 512 | | Strip New Lines | Remove newlines from input before embedding | true |

Troubleshooting

See docs/TOUBLESHOOTING.md

GitHub Enterprise Support

Set the Enterprise URL field in your credential to your GitHub Enterprise domain (without https://), for example:

company.ghe.com

The chat model node will automatically route requests to https://copilot-api.company.ghe.com. The embeddings node always uses the public GitHub Models API.

Credits

• Original package by @kk17: n8n-nodes-github-copilot-chat-models
• Embeddings node and fork by @lxgnzlz05

License

MIT