openai-custom
v0.10.9
Published
Activepieces piece for OpenAI with custom API URL and self-hosted endpoint support
Maintainers
Readme
OpenAI (Custom) Activepieces Piece (openai-custom)
An Activepieces piece for OpenAI with Custom API Base URL support. This piece allows you to connect not only to OpenAI's official API, but also to any OpenAI-compatible self-hosted or proxy endpoints (such as Ollama, OpenRouter, vLLM, LiteLLM, LM Studio, Groq, FastChat, and Azure OpenAI proxies).
Features
- Custom Base URL Support: Specify any custom endpoint (e.g.
http://localhost:11434/v1,https://openrouter.ai/api/v1, or your private corporate proxy). Defaults tohttps://api.openai.com/v1. - All 18 OpenAI Actions:
- Ask ChatGPT / Send Prompt
- Ask Assistant (with thread memory support)
- Vision Prompt (GPT-4o image understanding)
- Extract Structured Data
- Classify Text
- Analyze Sentiment
- Create Embeddings & Search Embeddings
- Generate & Edit Images
- Text-to-Speech (TTS)
- Transcribe & Translate Audio (Whisper)
- File Management (Upload, List, Find, Delete)
- List Models
- Custom API Call Action
- Standalone Package: Fully self-contained npm package ready to be published and installed in any Activepieces instance.
Installation in Activepieces
- Go to your Activepieces dashboard.
- Navigate to Pieces > Install Piece.
- Enter
openai-custom(or your npm package name) and the version.
Authentication
When configuring the piece connection in Activepieces:
- API Key (Required): Your OpenAI API key or compatible service key (e.g.
sk-...orollama). - Base URL (Optional): The base URL for the API endpoint (defaults to
https://api.openai.com/v1).
Build & Development
npm install
npm run buildPublishing to NPM
npm login
npm publish --access public