Advancing language models and natural language understanding.
Recently active
I’m interested in hearing from everyone here about which models you’re successfully using with your Metis device and Voyager SDK, and which ones you’d like to see being officially supported.This could be really useful info in terms of prioritisation when it comes to adding more support in the model zoo.
I just built a quick demo showing the Llama 3.2B chatbot running on our Metis® platform, totally offline. This model packs 3 billion parameters and runs smoothly on both a standard Lenovo P360 with our PCIe card and even on an Arduino-based dev board (Portenta X8).We hit 6+ tokens/sec with a single core – which means real-time chat. Perfect for smart customer support bot, digital concierge systems, any edge AI assistant application really, all running fully on-device. No cloud needed.Check out the video and let me know what you think. Any projects you can think of where you could use a self-contained, power-efficient, offline AI chatbot like this?
how let run a LLM learning process driven by system ram and cpu model outsource to metis device?or better say how to load 4,7GB into metis m.2 with 1GB
May 4th is almost upon us, so naturally we’re thinking about AI… in space. More specifically: the AI and droid legends of the Star Wars universe.From multilingual protocol units to rogue assassin bots with surprising moral compasses, Star Wars has given us some of the most iconic (and occasionally chaotic) artificial intelligences in sci-fi.So here’s the question...
This new survey suggests people (Apple fans, that is) are very open to the idea of paying a subscription for “Apple Intelligence” features. Which raises the question as to whether AI tools are going to become just another line on your monthly bill.We’ve moved quickly from “AI is cool” to “you might need it just to keep up.” And if it’s locked behind paywalls? That could become a problem once it’s a necessity rather than a nice-to-have.Curious what you all think – will AI be a must-have, or a paid perk for you?https://www.macrumors.com/2025/04/22/apple-intelligence-payment-survey/
There’s a lot of new hype everyday around new models that can do language + images + audio etc all at once. There’s a convenience to it, but is it the best way? Can’t help but think ”jack of all trades” when a new one is announced.I suppose it could be because of the chase towards the first AGI, but if you’re building an AI tool or platform, don’t you want a specialized model rather than a general one?Would be interested to hear if anyone here has actually built something on top of this — or tried and hit a wall.
Already have an account? Login
No account yet? Create an account
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.