Opera Browser Experiments With On-Device LLM Support
Web browser Opera has begun testing on-device support for large language models (LLMs). This will be the first built-in local LLM control for a browser, providing access to generative AI features without requiring the cloud to process the data, and serves as an alternative to the Aria generative AI assistant created by Opera.
Local LLM
The locally supported generative AI models include Google’s Gemma, Mistral AI’s Mixtral, Meta’s Llama, and Vicuna, a model based on Llama 2. Using local LLMs keeps users’ data locally on their devices, allowing them to utilize generative AI without the need to send information to a server.
The new feature is part of Opera’s AI Feature Drops Program. Those interested in trying it out need an updated version of Opera Developer to download their chosen LLM, which may require anywhere from two to ten gigabytes of storage. The local LLM will then be used instead of Aria until the user decides to switch back.
“AI is moving fast and so are we. We’ve started the AI Feature Drops Program to allow people to test our newest AI explorations that either will or won’t make it to the official version of Opera One,” Opera executive vice president Krystian Kolondra explained. “We are excited to let our most engaged users test and share their feedback and suggestions with us.”
On-device LLM support is an outgrowth of Opera’s ongoing AI development, starting with the AI-centered Opera One browser’s release last year. The company has quickly adopted ever more generative AI tools. Aria is the crown jewel of that strategy, acting as an assistant to look up information, summarize website content, and compose text. Later upgrades enhanced Aria’s writing ability and even enabled users to teach Aria to write in their style by submitting examples of their own writing.
Follow @voicebotaiFollow @erichschwartz
Opera Teaches Generative AI Assistant Aria to Mimic Your Writing Style
Opera Web Browser Launches Generative AI Features and ChatGPT Access
