I can run a 35b-3a AI model, on a single videocard, and have it translate a Japanese game within a day. A year ago, that would have required much more hardware, and would have taken longer. Your position is bad, because it doesn’t acknowledge a developing reality, and more importantly, surrenders the future to jerks by default.
I can run a 35b-3a AI model, on a single videocard, and have it translate a Japanese game within a day. A year ago, that would have required much more hardware, and would have taken longer. Your position is bad, because it doesn’t acknowledge a developing reality, and more importantly, surrenders the future to jerks by default.