TensorART is a hybrid API for tensor.art. It is able to communicate with Tensor using web-scraping.. It is able to access most used or useful ai features, such as text2img, img2img , upscale, and a lot more.

Warning

This project is probably against TensorART TOS. Use at your own risks.

Installation

• Install using pip (python 3.11 or higher required, recommended version):

pip install --upgrade TensorART

• Or from this repo to get the latest fixes/features:

pip install --upgrade git+https://github.com/XZeipher/TensorAI.git

• Or, for python 3.10 and higher.

pip install --upgrade git+https://github.com/XZeipher/TensorAI.git@py-3.10

• Or, for even lower python versions, there is automatic branch with several features removed (might be unstable).

pip install --upgrade git+https://github.com/XZeipher/TensorAI.git@compat

Quickstart

Note

You can find the docs on this project here.

from TensorART import TensorClient
import aiohttp,aiofiles

# Initialise a client
client = TensorClient()

# Create ai image
image_url = await client.create(model_id=1, prompt="blonde girl sitting in garden")

# Save image binary data
async with aiohttp.ClientSession() as session:
    async with session.get(image_url['url']) as response:
        async with aiofiles.open('image.png', mode='wb') as f:
            await f.write(await response.read())

# image will be saved as image.png

Note

This repository is initiated and maintained by XZeipher

License

TensorART uses GPLv3. See the LICENSE file.

Contributing

Feel free to contribute to this project by submitting feature requests, issues, bugs, or whatever.