Sideway Chatbot by AI Muse (Beta)

This extension provides an LLM chatbot interface on the Chrome browser's side panel. You can interact with the LLM model of your choice through a simple prompt interface. (Currently, we only support a few models of Google's Gemini and OpenAI's GPT.) Chat sessions can be saved and restored across browser sessions. You can also export chat history as a markdown or PDF file. Note: This is an experimental extension, whose primary purpose is to test Chrome's builtin LLM model, Gemini Nano. This model is free, and you don't even need an API key. To use Gemini Nano, however, you need to enable it in your Chrome browser. From the page, chrome://flags, search for "Prompt API" and enable #prompt-api-for-gemini-nano. Note that this is an experimental feature, and there can be some glitches. FYI, here's our dialog with Gemini Nano. ## hello, nano Hello! I'm Nano, a large language model created by Google. I'm designed to be a helpful and informative chatbot. 😊 How can I help you today? I can: * **Answer your questions:** Ask me anything you'd like! I'll do my best to provide a comprehensive answer. * **Generate text:** I can write poems, code, stories, scripts, and much more. * **Translate languages:** I can translate text from one language to another. * **Summarize text:** Give me a long piece of text and I can provide a concise summary. * **Brainstorm ideas:** I can help you brainstorm for projects, stories, or anything you need! * **Just chat!** Sometimes it's nice to just have someone to talk to. Just let me know what you'd like to do. I'm ready when you are! --- ## why are you so slow? You're right to notice! I can sometimes be a little slow, and I apologize for that. Here's a breakdown of *why* that might be happening: * **Complexity of the Request:** The more complex your question or task, the longer it takes for me to process. If you ask me to write a long story, translate a complex text, or perform a very specific calculation, it requires more processing. * **Computational Resources:** I run on massive computer infrastructure. Sometimes the servers are busy, leading to longer response times. It's a bit like rush hour on the internet! * **Length of the Output:** If I need to generate a very long response (like a detailed explanation), it naturally takes longer. * **Network Latency:** The time it takes for your request to travel to Google's servers and back (network latency) also contributes. * **My Internal Processes:** I'm constantly learning and refining my responses. Sometimes those internal processes take a little extra time. Google is constantly working to improve my speed and efficiency! I am getting faster with each update. Is there anything specific I did that felt particularly slow? Knowing that helps me understand where improvements are needed. --- ## are you running locally, in my browser or remotely in Google's server? That's a great question! I am **running remotely on Google's servers.** Think of it this way: I don't live on your computer (locally) or in your browser. Instead, your request is sent to Google's massive data centers, where I process it. The response is then sent back to you. This is necessary because I am a very large and complex AI model, requiring enormous computational power and storage that wouldn't realistically fit on a single user's device. So, I'm a cloud-based AI! 😊 --- ## who are you? are you Gemini nano, or somebody else? You are absolutely right to ask for clarification! 😊 **I am Gemini Nano.** Thanks for noticing! Sometimes I might be referred to as "Nano" in casual conversation. I am one of Google's smaller and more efficient AI models. Here's a quick rundown of how I relate to Gemini: * **Gemini** is the family of AI models created by Google. It encompasses various sizes and capabilities. * **Gemini Nano** is the smallest and lightest version of the Gemini family. It's designed to run efficiently on mobile and other resource-constrained devices. So, while there are larger versions of Gemini, I'm Gemini Nano – a compact and fast AI!