Scope4 Beta APIs

Updated May 13, 2025

Welcome to the Scope4 Beta APIs page! Here you'll find information about new APIs currently in the beta testing phase. These APIs offer cutting-edge functionalities to enhance your sustainability analyses. As beta features, they are subject to change and refinement based on user feedback.

ScopePDF API

The ScopePDF API is designed to extract structured data from PDF documents, focusing on sustainability-related information. It currently has two primary endpoints in beta:

1. Bill of Materials (BoM) Data Extraction

This endpoint is dedicated to extracting detailed information from Bill of Materials (BoM) documents. It helps automate the process of identifying components, materials, quantities, units, and sourcing information crucial for Life Cycle Assessments (LCA) and carbon footprint calculations.

Key Features:

  • Automated Data Extraction: Upload your BoM PDF, and the API uses advanced capabilities to parse and extract relevant data.
  • Structured Output: Receive data in a clean JSON format, listing materials with their respective names, quantities, units, and associated year and geography if available.
  • Focus on Sustainability: Tailored to capture information pertinent to environmental impact analysis.

Availability & Testing:

  • This endpoint is currently available for testing.
  • Access the Tutorial: You can test the ScopePDF BoM endpoint using the ScopePDF API Tutorial Colab Notebook.
  • Credentials: To use this API, you'll need a username and an API key. These can be obtained by submitting a request via the Builder Form or the Enterprise Form on scope4.dev (the same process as for the ScopeGreen API).

How it Works (Simplified):

  1. Authentication: The user provides their username and API key.
  2. PDF Upload: The user uploads a BoM PDF file through the provided interface (e.g., Colab notebook).
  3. Backend Processing: The PDF is sent to a secure backend where it's processed.
  4. Data Extraction: The system analyzes the PDF content based on a predefined schema to identify materials, quantities, units, etc.
  5. JSON Output: The extracted data is returned to the user in a structured JSON format, specifically designed for ease of use in further sustainability calculations.

2. EPD Data Extraction & Quality Evaluation

This upcoming endpoint will focus on extracting data from Environmental Product Declarations (EPDs) and evaluating the quality of the extracted data.

Key Features (Planned):

  • Automated data extraction from EPD PDF documents.
  • Evaluation of data quality and completeness based on EPD standards.
  • Structured output of key EPD metrics.

Availability:

  • Details and testing availability for this endpoint will be announced soon.

ScopeLLM API

The ScopeLLM API is a specialized service for calculating the environmental footprint associated with the usage of Large Language Models (LLMs). As LLMs become more prevalent, understanding their energy consumption, water usage, and carbon emissions is increasingly important.

Key Features:

  • Impact Calculation: Automatically provides the relative energy, water, and carbon footprint based on LLM usage.
  • Simple Inputs: Requires the model name and the number of tokens processed as inputs.
  • Focus on LLM Sustainability: Helps users and organizations assess and report the environmental impact of their AI operations.

Inputs:

  • model_name: The specific Large Language Model used (e.g., "gpt-4", "claude-3-opus").
  • token_numbers: The number of input and output tokens processed by the LLM.

Output:

  • The API returns the relative impact values for energy consumption, water usage, and carbon footprint.

Availability:

  • The ScopeLLM API is currently in the beta phase. Further details, including testing access and specific output formats, will be made available soon.

Feedback

Your feedback on these beta APIs is invaluable to us. Please share your experiences, suggestions, or any issues encountered by contacting us at tommaso@scope4.dev.