There are times when local AI is the topic again. This time too. My computer isn’t that good, but I’ve tried it once. In conclusion, I don’t really want to recommend it yet.
If you’re thinking about buying a new PC for local AI, there are some things you should ask yourself before deciding.
Personally, if you want to make money with AI without any business (or subsidiary) infrastructure, I would recommend that you never start with local AI. The reason is simple. The model continues to improve, but my hardware remains the same.
On the other hand, when you pay a subscription and use a cloud-based model, you get better performance without doing anything. My results naturally get better together.
But there are exceptions.If you already have your own workflow or system built, the story is different.
For example, let’s say that you’re already running a service or system and it’s going back on a fairly large scale. In this case, it can be more efficient to monitor or process something 24 hours a day with a local AI than to pay an API each time and send a request to the cloud, because it’s more cost-effective to turn repeated and normalized tasks from local to local, especially when it comes to security.
The benefits of local AI are clear if you don’t have to exit data or have to repeat the same task hundreds of times a day.
We do not recommend preparing from the local AI with the idea of “Let’s try something with AI” when there is no basis yet, but if there are workflows or business items that are already rolling around, and you can give the local AI a repetitive and specific role in it, then I think it’s worth investing.
Loading comments.