Not at all, it's more than enough for a large range of tasks. As for slow, that's just a function of how much compute you throw at it, which you actually control unlike with closed weights models.