labview-llm-libraries

This repository contains the LabVIEW LLM Libraries, including packages and examples that enable developers to seamlessly integrate AI capabilities into their LabVIEW applications. It provides foundational building blocks for AI application development in LabVIEW, allowing developers to easily compose custom AI solutions. The focus is on maximizing modularity and composability, empowering users to build tailored solutions with ease.

What can you do with these libraries?

1. Interact with your favorite large language models (LLMs) using LabVIEW in 5 min.
2. Build your own simple RAG solution (Chat Bot) in LabVIEW in 20 min.
3. Vision is to provide integration to data analysis and visualization modules (SAAS)

Key Features of the Libraries

1. Interact with large language models (LLMs) be it online or offline models. The libraries provide pre-built support for popular models. But you can also use the abstraction to build interface supports to your own models, if needed.
2. Use the libraries to build your a RAG (Retrieval-Augmented Generation) system, which combines the power of LLMs with a retrieval mechanism to enhance the quality and relevance of generated content.
3. Interact with the vector databases for efficient storage and retrieval of embeddings. The libraries provide seamless integration with popular vector databases, but also allow for custom implementations if required.
4. Utilize the libraries to build your own data processing pipelines for tasks such as data extraction, chunking, embedding.
5. Interfaces to do data analysis and dynamic visualization using LabVIEW

Getting Started

1. Download and Install: Access the latest version of the LabVIEW LLMs Package from the Releases section of this repository.
2. Explore Examples: Use the examples provided in this repository to perform data vectorization and create custom chat applications.
3. Supported LabVIEW Version: >=2020

Models supported

Model Name	Provider	Support Available	Online/Offline
gpt-4o-mini	OpenAI	Yes	Online
gpt-4o	OpenAI	Yes	Online
gpt-4-turbo	OpenAI	Yes	Online
gpt-4	OpenAI	Yes	Online
gpt-3.5-turbo	OpenAI	Yes	Online
gpt-4o-mini	AzureOpenAI	Yes	Online
gpt-4o	AzureOpenAI	Yes	Online
gpt-4	AzureOpenAI	Yes	Online
gpt-3.5-turbo	AzureOpenAI	Yes	Online
Claude 3.5 Sonnet	Anthropic	Yes	Online
Claude 3 Opus	Anthropic	Yes	Online
Claude 3 Sonnet	Anthropic	Yes	Online
Claude 3 Haiku	Anthropic	Yes	Online
Llama 3.1 8B	Ollama	Yes	Offline
Llama 3 8B	Ollama	Yes	Offline
Phi 3 Mini 3.8B	Ollama	Yes	Offline
Gemma 2 2B	Ollama	Yes	Offline
Mistral 7B	Ollama	Yes	Offline
tinyllama	Ollama	Yes	Offline

Embedding Models supported

Model Name	Provider	Support Available	Online/Offline
text-embedding-ada-002	OpenAI	Yes	Online
text-embedding-3-large	OpenAI	Yes	Online
text-embedding-3-small	OpenAI	Yes	Online
nomic-embed-text	Ollama	Yes	Offline
mxbai-embed-large	Ollama	Yes	Offline

Vector Databases Supported

Database Name	Provider	Support Available	Online/Offline	Setup Instruction
Weviate	Weaviate	No	Offline
Qdrant	Qdrant	Yes	Offline	qdrant setup

Support

If you encounter any issues while using the LabVIEW LLM Libraries or have suggestions or feature recommendations, please share your feedback in the Issues section of this GitHub repository.