4 ms·
It’s all just tokens to the model. Whether it’s thinking tokens or output tokens, they take the same amount of computation to produce. The only difference is wh
by brokencode 2mo ago
It’s all just tokens to the model. Whether it’s thinking tokens or output tokens, they take the same amount of computation to produce. The only difference is whether the token is displayed to the user.