rqle-ai-langchain-util 0.4.1

Library facilitating the integration of different LLM providers in LangChain (e.g. `ollama`, `Google Gemini`, etc).

LangChain Utility

• Description
• Documentation
• Development
  • Requirements
  • Environment Variables
  • How to prepare the environment
• About RQle.AI
• Disclaimer

Description

LangChain is a framework for developing applications powered by language models. It enables application that:

• Are context aware; i.e. connect a language model to sources context
• Reason; i.e. rely on a language model to reason

This library offers integration with the following Generative AI servers:

• AWS Bedrock
• Azure OpenAI
• Google AI Studio
• ollama
• Oracle Cloud Infrastructure AI

This library extends LangChain by providing facilities to define configurations, templates for execution of LLMs.

Documentation

To get started refer to the documentation.

Development

Requirements

• Git
• llm-guard >= 0.3
• langchain >= 0.3
• traceloop-sdk >= 0.33
• Python = 3.11
• Poetry >= 1.8.4
• streamlit >= 1.32

Environment Variables

Name	Description	Default
LOG_DIR	Location of the logging files	logs/
LOG_LEVEL	Logging level to be applied during execution	INFO
AUDIO_TMP_FOLDER	Location of temporary audio files when transcribing videos	tmp/audio/
PROMPT_CONFIG_FOLDER	Location of the Prompt configuration for execution	prompt_configs/

How to prepare the environment

• Install dependencies

  poetry install
  

  ---

  NOTE To update dependencies, it may be needed to run the following command prior to installing the packages:

  poetry lock
  

  ---

• Test unit test coverage for the project

  poetry run coverage run -m pytest && poetry run coverage report -m
  

  Note Report is only generated if all unit test have completed successfully.

About RQle.AI

RQle.AI believes in the transformative potential of Generative AI. More specifically, it focuses on showcasing real-world applications of how Generative AI can empower individuals and organizations worldwide in addressing customers' "job to be done" problems and create value for them.

Disclaimer

This library and its use of Large Language Models (LLMs) are subject to the following disclaimers:

• LLMs are still under development and may generate inaccurate, incomplete, or biased output;
• LLMs can inherit and reflect biases present in their training data;
• Developers of this library are not liable for any damages or losses arising from its use;
• You are responsible for using the library and LLMs in an ethical and responsible manner.

By using this library, you acknowledge and agree to these disclaimers and limitations.

---

  RQle.AI - 2024