The tok/s stat is interesting. Since the dominant constraint on inference speed is hardware, it suggests X purchased far more compute than was really needed to serve the demand for their models.
Expensive miscalculation.
The tok/s stat is interesting. Since the dominant constraint on inference speed is hardware, it suggests X purchased far more compute than was really needed to serve the demand for their models.
Expensive miscalculation.
Didn't a bunch of hardware that was destined for Tesla get redirected to xAI? I'm sure I remember something like that.
Yep! Why his shareholders in Tesla abide by this kind of thing is beyond me, but he often mixes resources from completely unrelated companies: https://www.cnbc.com/amp/2024/06/04/elon-musk-told-nvidia-to...