Skip to main content

Coconut AI

Project description

coconut-ai

About

Coconut AI is a Python library that allows you to use AUTOMATIC1111/Stable Diffusion Web API and LLama 2, to generate images from text, images to images or text to text.

The library is available on PyPy: https://pypi.org/project/coconut-ai/.

Getting Started

Prerequisites

Installation

pipenv install coconut-ai

Stable Diffusion Guide

For a complete example, see the coconut-ai-example repository.

Install Stable Diffusion Web UI

AUTOMATIC1111/Stable Diffusion Web UI installation guide

# Clone the AUTOMATIC1111/Stable Diffusion Web UI repository
git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git

# Go to AUTOMATIC1111/Stable Diffusion Web UI source code
cd stable-diffusion-webui

# For Windows users, run the following command for starting the server
./webui-user.bat --api --nowebui

# For Linux/MacOS users, run the following command for starting the server
./webui.sh --api --nowebui

Stable Diffusion Usage

Note: Before using the library, you must start the Stable Diffusion Web UI server.

# Go to your Python project/source code
cd ../python-project
pipenv install coconut-ai

Create a file named main.py with the following content (minimal example):

import os

import coconut_ai

if __name__ == '__main__':
    print("Coconut AI")
    output_path = os.path.abspath("output.png")
    os.makedirs(os.path.dirname(output_path), exist_ok=True)
    coconut_ai.text_to_image({
        "input_text": "Coconut",
        "output_path": output_path,
        "steps": 5
    })

Run the script:

pipenv run python main.py

LLama 2 Guide

Install LLama 2 Model

For a complete example, see the coconut-ai-example repository.

Download the model llama-2-7b.Q5_K_M.gguf manually.

All available models can be found on the Hugging Face website.

LLama 2 Usage

# Go to your Python project/source code
cd ../python-project
pipenv install coconut-ai

Create a file named main.py with the following content (minimal example):

import os

import coconut_ai

if __name__ == '__main__':
    print("Coconut AI")
    # Replace with the model path for the llama-2-7b.Q5_K_M.gguf model downloaded manually
    model_path = os.path.abspath("data/llama-2-7b.Q5_K_M.gguf")
    print(coconut_ai.text_to_text({
        'input_text': "Generate a list of 5 funny dog names.",
        'max_tokens': 1000,
        'model_path': model_path
    }))

Run the script:

pipenv run python main.py

💡 Contributing

Anyone can help to improve the project, submit a Feature Request, a bug report or even correct a simple spelling mistake.

The steps to contribute can be found in the CONTRIBUTING.md file.

📄 License

MIT

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

coconut-ai-1.0.0.tar.gz (4.3 kB view details)

Uploaded Source

File details

Details for the file coconut-ai-1.0.0.tar.gz.

File metadata

  • Download URL: coconut-ai-1.0.0.tar.gz
  • Upload date:
  • Size: 4.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/4.0.2 CPython/3.11.7

File hashes

Hashes for coconut-ai-1.0.0.tar.gz
Algorithm Hash digest
SHA256 726f23d7a58a7bc5fec806374a31676810dd2edd087e30a8451f51faa224c249
MD5 ec39f95553a3ecc107d1102a4c1b10ae
BLAKE2b-256 85051dcccd1c44c0471935fbad827a949b8bbf8c0d90e5fbf74f418a54dbc114

See more details on using hashes here.

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