The biggest, baddest Llama 3.1 405 billion parameter model is live on Venice.ai for Pro users only.

Use this model if you want the most intelligent results, however....

- 405b is not connected to web, so stick with web-version of Nous Theta for any real-time questions

- Inference is slow relative to smaller models, future optimizations possible, but we wanted to be live ASAP

h/t to our GPU partner @hyperbolic for the fast work on this 🚀 and of course to Meta for open-sourcing a true state of the art frontier model.

The revolution will not be centralized.