the most impressive thing about this isn't that some random japanese company created a mythos-level model - its *how* they did it:
-> their ai model isn't actually a model, it's an API that calls *other models* (e.g. chatgpt, claude, their own)
-> their orchestrator selects different models to do different parts of your prompt. if a cheaper model can be used they'll do that. thats how they cut costs.
-> if a task is challenging then they'll use a frontier model (e.g. claude) to design a solution, then use a cheaper model to build it.
point is - frontier ai capability is no longer solely dependent on how good the model weights are, its how MANY model instances you can get to debate and come up with an answer between themselves
more models going back and forth = better cheaper answer.
we're moving from mono-model to multi-model
A place where everyone you ever loved or cared about will be playing in a beach colored by the evening sun. A place where you can see yourself from afar in those memories, relive how you felt in that moment and be happy that you ever got the chance to make them.
After some time thinking about why nostalgia is so addictive. I came to the conclusion that it is because it crystalizes the best memories that you ever had in an inexorable and eternal moment in time.
You get the sadness because it is unreachable, and you will not touch it again in this life. But you feel good because it reminds you that there is a place somewhere where it never fades.