Press Space to continue
This approach reduces inference costs by routing most requests
Press Space to continue
I think the ideal model router actually is just starting with a cheap model and teaching it to escalate as needed + using a more powerful model as a “checker” that also can escalate
@theoQuoted post unavailable hereOpen on Xopen on x · 3 viewsViews and clicks count once per visitor per day. Clicks are opens of this post on X.