aimped

aimped

Welcome to Aimped Inc.: an LLMs-in-the-loop AI Platform.

Aimped is a distinctive AI platform offering model inference and development, harnessing the power of LLMs. Our innovative approach ensures quick, cost-effective, and efficient model training and evaluation. Additionally, we provide user-friendly UI-based AI applications and LLM Agents, making advanced AI accessible and easy to use.

Moreover, we will soon allow model uploads and enable users to earn money. This will encourage the open-source community to share models, reducing redundant training and computing costs while fostering collaboration and innovation.

aimped is a unique python library that provides classes and functions for only exclusively business-tailored AI-based models. In this version, we provide the following features: API service, Sound processing tools and functions, NLP tools and functions, and a pipeline class for NLP tasks.

Installation

pip install aimped

API Usage

Configuration of the Library

from aimped.services.api import AimpedAPI

# Create new instance Aimped
user_key = ''  # user_key received from A3M.
user_secret = ''  # user_secret received from A3M.
BASE_URL = 'https://aimped.ai'  # Aimped domain url

api_service = AimpedAPI(user_key, user_secret, {
  base_url: BASE_URL
})

Preparation of the model input data

model_id = ""   # ID of the model run. The model ID is available on the model description page under API usage. 
payload = {...}  # Model input examples (payload) are available in the api usage tab on the Model description page. 

Usage of API Function

result = api_service.run_mode(model_id, input_data)

Usage of API Callback Function

# return callback function

def callback(event, message, time, data=None):
    if event == 'start':
        print(f'Start event at {time}: {message}')
    elif event == 'proccess':
        print(f'Progress event at {time}: {message}')
    elif event == 'error':
        print(f'Error event at {time}: {message}')
    elif event == 'end':
        print(f'End event at {time}: {message}. Data: {data}')

result = api_service.run_model_callback(model_id, payload, callback)

Usage of API File Upload

Some of the models supports file inputs. These inputs are accepted as URIs. Here is the usage of API for file uploads.

input = api_service.file_upload(
    model_id,
    '/Users/joe/Downloads/xyz.pdf'  # sample file path to upload
    )

Usage of API File Download

Some of the models supports file outputs as result. These outputs are created as URIs. Here is the usage of API for file downloads.

  output_file = api_service.file_download_and_save(
    'input/application/model_{{modelId}}/user_{{userId}}/file_name',  # URI of the model output file in the result
    '/Users/joe/Downloads/123_file_name'  # sample local file path to save
    )

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.

License

MIT