Hi, Nayan from Webbotify here. I'm excited to share that we now support Meta Llama 3, the most capable open-source model on Webbotify.
There are two variants available:
- Llama 3 8B - This is the smaller variant, trained on 8 billion parameters, and its performance is very close to GPT-3.5 in most use cases.
- Llama 3 70B - This is the larger variant, trained on 70 billion parameters, with great performance and speed, which we found comparable to GPT-4.
To check out this model today, go to the settings page and choose the model section.
