new2시간 전· 1 sources
It agreed with the reference 100% of the time. It was right 75% of the time.
Why it matters
Swap a cheap model in behind an expensive one and the obvious way to check it is shadow traffic: send the same request to both, compare the answers, count how often they agree. High agreement, ship it...
1
Sources
+1
24h
—
Growth
1d
Active