From the lab

Notes, evals and release write-ups

Everything we publish about the model — including the parts we would rather not.

How to read an agent leaderboard honestly

Error terms, session counts and why a 4.8% net improvement is not a rounding accident.

The first Turkish foundation model to ship weights, a datasheet and a leaderboard entry at the same time.

Agglutination, vowel harmony and the quiet tax that general-purpose tokenizers put on every Turkish sentence.