Skip to main content

Generate Data from LLM easily using LLMKit

Reason this release was yanked:

Bug fix for examples

Project description

LLMKit - Language Model Kit

LLMKit is a Python package that simplifies the interaction with Language Models (LLMs) and allows you to generate text data with ease.

Features

  • Simple Interface: LLMKit provides a straightforward interface to interact with various language models.

  • Custom Transformers: You can create custom transformers to format input data into prompts and transform LLM-generated text into the desired data format.

  • Model Abstraction: LLMKit abstracts the details of model selection and provider handling.

Installation

You can install LLMKit using pip:

pip install llm-stack-kit

Usage

from llmkit import LLMKit

llm_kit = LLMKit()

input_data = "Translate the following English text to French: 'Hello, World!'"

# Generate text using the default transformers
generated_text = llm_kit.generate_data(input_data)

print("Generated Text:", generated_text)

Configuration

For more advanced usage and customization, you can create your own transformers and specify model providers. Only two model providers are supported for now: "openai" and "cloudflare". You need to provide a .env file with either OPENAI_API_KEY for OpenAI or CF_API_KEY and CF_ACCOUNT_ID for Cloudflare.

See sample custom transformer for how to write a custom transformer.

Documentation

Comprehensive documentation will be made available soon.

Contributing

We welcome contributions! Our contributions guideline will be added soon.

License

LLMKit is distributed under the MIT License. See LICENCE for details.

Contact

If you have any questions or suggestions, feel free to reach out to us at useentropy@gmail.com.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

llm_stack_kit-0.1.2.tar.gz (7.0 kB view hashes)

Uploaded source

Built Distribution

llm_stack_kit-0.1.2-py3-none-any.whl (9.5 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page