Vision-Claw
Meta's default Llama experience did not meet the bar for accuracy, speed, or depth of knowledge.
Core trick
Routes live camera frames and voice from smart glasses into Gemini Live, which forwards requests to a local agent that executes actions.
Description
Meta's default Llama experience did not meet the bar for accuracy, speed, or depth of knowledge. I wanted my glasses to feel less like a novelty assistant and more like an always-available interface to the best models and the tools I already use.
So I reverse-engineered the integration path and brought Gemini Live into the glasses experience. Then I connected my Hermes agent, turning a voice-and-vision assistant into an interface that can understand context and take action.
User can Use gemma models locally for privacy
README excerpt
# Super Meta ### A hands-free, Gemini-powered copilot for Ray-Ban Meta glasses. **See the moment. Ask naturally. Act through your agent.** [](#requirements) [](#architecture) [](#the-experience) [](#requirements) Built for the [Google DeepMind Bangalore Hackathon](https://cerebralvalley.ai/e/google-deepmind-bangalore-hackathon/details), Super Meta turns the glasses camera, microphone, and a personal agent into one continuous interface. Say **"Hey Gemini"**, look at the world, and keep moving. > `Super Meta` is a development codename. "Meta" is a trademark of Meta Platforms; rename the project before public distribution. ## Story Meta's default Llama experience did not meet the bar for accuracy, speed, or depth of knowledge. I wanted my glasses to feel less like a novelty assistant and more like an always-available interface to the best models and the tools I already use. So I reverse-engineered the integration path and brought **Gemini Live** into the glasses experience. Then I connected **my Hermes agent**, turning a voice-and-vision assistant int