llmx 0.0.6a0

LLMX: A library for LLM Text Generation

LLMX - An API for Language Models

A simple python package that provides a unified interface to several LLM providers [ OpenAI (default), PaLM, Cohere and local HuggingFace Models ].

There is nothing special about this library, but some of the requirements I needed when I startec building this (that other libraries did not have):

• Unified Model Interface: Single interface to create LLM text generators with support for multiple LLM providers.

from llmx import  text_generator as generator

openai_generator = generator(provider="openai")
palm_generator = generator(provider="google") # or palm
cohere_generator = generator(provider="cohere") # or palm
hf_generator = generator(provider="huggingface") # run locally

• Unified Messaging Interface. Standardizes on the OpenAI ChatML format. For example, the standard prompt sent a model is formatted as an array of objects, where each object has a role (system, user, or assistant) and content of the form. A single request is list one only one message (e.g., write code to plot a cosine wave signal). A conversation is a list of messages e.g. write code for x, update the axis to y, etc. For all models.

messages = [
    {"role": "user", "content": "You are a helpful assistant that can explain concepts clearly to a 6 year old child."},
    {"role": "user", "content": "What is  gravity?"}
]

• Good Utils (e.g., Caching etc): E.g. being able to use caching for faster responses. General policy is that cache is used if config (including messages) is the same. If you want to force a new response, set use_cache=False in the generate call.

response = gen.generate(messages=messages, config=TextGeneratorConfig(n=1, use_cache=True))

Output looks like

TextGenerationResponse(text=[Message(role='assistant', content="Gravity is like a magical force that pulls things towards each other. It's what keeps us on the ground and stops us from floating away into space. ... ")], config=TextGenerationConfig(n=1, temperature=0.1, max_tokens=8147, top_p=1.0, top_k=50, frequency_penalty=0.0, presence_penalty=0.0, model_type='openai', model='gpt-4', stop=None), logprobs=[], usage={'prompt_tokens': 34, 'completion_tokens': 69, 'total_tokens': 103})

Are there other libraries that do things like this really well? Yes! I'd recommend looking at guidance which does a lot more. Interested in optimized inference? Try somthing like vllm.

Installation

Install from pypi. Please use python3.9 or higher.

pip install llmx

Install in development mode

git clone
cd llmx
pip install -e .

Note that you may want to use the latest version of pip to install this package. python3 -m pip install --upgrade pip

Usage

Set your api keys first for each service.

export OPENAI_API_KEY=<your key>
export PALM_API_KEY=<your key>
export COHERE_API_KEY=<your key>

# for palm ..
export PALM_SERVICE_ACCOUNT_KEY_FILE= <path to your service account key file>
export PALM_PROJECT_ID=<your gcp project id>
export PALM_PROJECT_LOCATION=<your project location>

from llmx import  text_generator as generator
from llmx.datamodel import TextGenerationConfig

messages =  messages = [
    {"role": "system", "content": "You are a helpful assistant that can explain concepts clearly to a 6 year old child."},
    {"role": "user", "content": "What is  gravity?"}
]

openai_gen = generator(provider="openai")
openai_config = TextGenerationConfig(model="gpt-4", max_tokens=50)
openai_response = openai_gen.generate(messages, config=openai_config, use_cache=True)
print(openai_response.text[0].content)

See the tutorial for more examples.

Current Work

• Supported models
  • OpenAI
  • PaLM
  • Cohere
  • HuggingFace (local)

Caveats

• Prompting. llmx makes some assumptions around how prompts are constructed e.g., how the chat message interface is assembled into a prompt for each model type. If your application or use case requires more control over the prompt, you may want to use a different library (ideally query the LLM models directly).
• Inference Optimization. This library is not really designed for speed, but more for rapid experimentation using multiple models. If you are looking for a library that is optimized for inference, I'd recommend looking at vllm or tgi

Citation

If you use this library in your work, please cite:

@software{victordibiallmx,
author = {Victor Dibia},
license = {MIT},
month =  {10},
title = {LLMX - An API for Language Models},
url = {https://github.com/victordibia/llmx},
year = {2023}
}