Why Did Google Split Its AI Chips Into 8i and 8t?

Amin Vahdat says Google split the line because, by 2026, inference and serving had grown large enough to justify a chip specialized for each job. About two years earlier the question was whether 2026 should bring one chip that does inference and training together, or two chips, one further specialized for inference and one for training. The analysis produced two chips: 8i for inference and 8t for training.

Specialization is a calculus. The more a chip is aimed at one workload, the less flexible it is, and the faster and more power-efficient it gets. A large fixed cost has to be covered by enough durable demand. A workload that disappears after a month or two is a narrow window, even when it is large for that short stretch. If inference had looked like 2 percent or 5 percent of the market, even a chip twice as fast might not have been worth a separate program. Serving looked like it might be 30, 40, 50, or 60 percent of the market over the chip's life, which made the split make sense.

Both chips can still run the other workload. If 8i had zero training performance and 8t had zero inference performance, Google would have had to predict, across a six-year hardware life, exactly how many of each to buy. Leftover capacity on either side can still do the other job.