| ▲ | Kuyawa 8 hours ago | |
Duh, China? | ||
| ▲ | verdverm 7 hours ago | parent [-] | |
We also see improvements like 7x between the current and last Google TPU generations, with different chips for training vs inference. | ||
| ▲ | Kuyawa 8 hours ago | |
Duh, China? | ||
| ▲ | verdverm 7 hours ago | parent [-] | |
We also see improvements like 7x between the current and last Google TPU generations, with different chips for training vs inference. | ||