New Mojo AI Programming Language with RWKV-LM? = Performance Boost?

[Peter2023AI]

I was checking into the new Mojo AI programming language, and the possible significant performance boost advantages it might provide to RWKV-LM. Has anyone considered checking into this for RWKV-LM?

Here are the steps according to Mojo’s website to start and finish a project using Mojo:

1. Installation: Download the Mojo distribution from their website and decompress the zip file. Make sure you have installed the required support packages: Tornado, TornadIO2, bcrypt, and wtforms. From the command line in the new folder run python setup.py install. This should install Mojo into your python path.

2. Create a project: To create your first app, you simply need to invoke the mojo_manager application. This will create your project folder with a typical structure that includes a project folder that contains a series of Apps. These apps are independent from one another and offer a way to group different functional areas of your app away into discrete units.

3. Develop your app: Each app has a series of base files that make your app work: models.py for model definition, ui_modules.py for UI Modules for Tornado, urls.py for mapping URLs to relevant Request Handlers, and views.py for request handlers that will manage the various app’s functions.

4. Test and deploy: Once you have developed your app, you can test it locally and then deploy it to a server.

• Video about Mojo: https://youtu.be/Z7BXmpzaiCg
• How-to: https://mojo.readthedocs.io/en/latest/docs/quickstart.html