large language model

“We're now using small language models and even large language models to power The avatars that are teaching about these particular subjects.”

“You may actually be better off starting with professional services, hardware, an existing business, and then asking, you know, Hey, How do I leverage this existing cash flow to do things better? The irony of it is, is your odds of success are probably higher. …”

“And in doing that, we've added dimensions that the semantic layer itself, even if you enrich its contextual capabilities, will not be able to do. Unless married with other technologies, which is what we're doing here.”

“We don't actually use the large language models today in our product. So this is more traditional machine learning, you know, from proprietary data like time series data.”

“So my suggestion usually to everybody is start with the biggest and you can test the smaller ones to see if they keep up or keep well enough for what you're trying to do. But if it really matters, start with the biggest and then use a tool like swept to speed …”

“I think that's a much more sort of robust and resistant to bad actor kind of design than having one giant LLM. You don't know how it operates because that's another problem. They're big black boxes and the value system was programmed in by five people at open …”

“Each individual entity, whether it's a large language model or human, doesn't need to be a super genius. You can just be kind of okay. But together in aggregate, if you coordinate them in the right way, the community itself will be smarter than anything on the…”

“So nothing relies on an LLM at the moment. Everything relies on our proprietary computer vision model, which was built on a neural network and shrunk down to operate entirely on device.”

“The world of, I'm going to take an LLM, I'm going to wrap it around some particular workflow or use case and think that builds a business. I mean, we, we've pretty much seen in the market that that's, there's not a lot of belief in that, right?”

“I think the only right solution in this case is, Even though you have a local model, or even though you're using the enterprise safe version of the model is just to eliminate the data completely. So if you don't have anything sensitive that goes to AI in the f…”