Back to the catalogue
Chat modelOpen weights
Llama 4 Maverick
Open-weights model you can also run on your own hardware. A cheap default for high-volume, low-stakes answering.
Good enough for the questions that make up most of a support inbox, at a fraction of a frontier model's cost. Run it as the answering model behind an embedded widget and keep the expensive model for escalations.
Because the weights are open, an Enterprise deployment can point Ragenta at its own endpoint and keep every token inside its own network.