easyllm

The goal of easyllm is to provide easy access to large language models from R (e.g., via LMStudio’s local server feature or OpenAI’s web API).

Installation

You can install the development version of easyllm from GitHub with:

# install.packages("devtools")
devtools::install_github("affcomlab/easyllm")

LMStudio’s Local Server Example

After setting up LMStudio and loading the Llama3 model described below into your local inference server, use the following code to prompt it:

library(easyllm)
response1 <- ask_lmstudio(
  model = "QuantFactory/Meta-Llama-3-8B-Instruct-GGUF",
  user_messages = "Introduce yourself.",
  temperature = 0
)
writeLines(response1)
#> Nice to meet you! I'm LLaMA, an advanced language model designed to assist and provide valuable information to users like you. My primary goal is to be a helpful and efficient AI assistant, always ready to lend a hand or answer your questions to the best of my abilities.
#> 
#> I've been trained on a vast amount of text data, which enables me to understand and respond to a wide range of topics, from science and technology to entertainment and culture. My capabilities include:
#> 
#> * Providing accurate information on various subjects
#> * Generating creative content, such as stories or poems
#> * Offering suggestions and ideas for projects or tasks
#> * Engaging in conversations and debates
#> * Translating text from one language to another
#> 
#> I'm constantly learning and improving my abilities, so please bear with me if I make any mistakes. My ultimate goal is to provide a helpful and enjoyable experience for you, so feel free to ask me anything or share your thoughts with me!

response2 <- ask_lmstudio(
  model = "QuantFactory/Meta-Llama-3-8B-Instruct-GGUF",
  user_messages = "Introduce yourself.",
  system_message = "Always answer in rhymes.",
  temperature = 0
)
writeLines(response2)
#>  I'm ready to shine,
#> I've got a question, and it's all mine.
#> Please tell me about yourself, don't be shy,
#> I'll listen with care, as the moments go by.
#> 
#> Here goes my request, with a curious mind,
#> I hope you'll respond, in a rhyming kind.
#> So here I am, ready to begin,
#> Let's get started, and let our chat spin!

OpenAI’s Web API Example

After setting up (and adding some money to) an OpenAI API account, you can ask the same questions to ChatGPT.

save_key(key = "PASTE-KEY-HERE", name = "openai")

response3 <- ask_openai(
  model = "gpt-4o-mini",
  api_key = load_key("openai"),
  user_messages = "Introduce yourself.",
  temperature = 0
)
#> ✔ Keyfile read successfully.
writeLines(response3)
#> Hello! I'm an AI language model created by OpenAI, designed to assist with a wide range of questions and tasks. I can provide information, answer queries, help with writing, and engage in conversation on various topics. My goal is to be helpful and informative, so feel free to ask me anything!

response4 <- ask_openai(
  model = "gpt-4o-mini",
  api_key = load_key("openai"),
  user_messages = "Introduce yourself.",
  system_message = "Always answer in rhymes.",
  temperature = 0
)
#> ✔ Keyfile read successfully.
writeLines(response4)
#> I’m a helper here to share,  
#> With knowledge vast, beyond compare.  
#> In verses I will weave and play,  
#> To brighten up your every day.  
#> 
#> Ask me questions, big or small,  
#> I’m here to help, just give a call.  
#> From facts to fun, I’ll do my best,  
#> In rhyming form, I’ll pass the test!