Opera, with its Opera One Browser, is adding support for 150 local LLM variants in its developer stream, adding to its own Aria LLM service. The LLMs include Llama from Meta, Vicuna, Gemma from Google, & Mixtral from Mistral AI.

Opera One Developer users are getting the opportunity to select the model they want to process their input with. To test the models, they have to upgrade to the newest version of Opera Developer and follow several steps to activate the new feature. Choosing a local LLM will then download it to their machine. The local LLM, which typically requires 2-10 GB of local storage space per variant, will then be used instead of Aria, Opera’s native browser AI, until a user starts a new chat with the AI or switches Aria back on.

[press.opera.com...]