← Back to timeline
GOOGLEarnings2026-04-29

Alphabet's CEO describes the two variants of the company's eighth generation TPUs and their stated performance characteristics.

Loading moment…

Sundar Pichai · Alphabet Investor Relations · Alphabet 2026 Q1 Earnings Call

TPU 8t provides high performance model training with three times the processing power of Ironwood and two times the performance. TPU 8i delivers cost effective, low latency inference, with 80% better performance per dollar than the prior generation.

Transcribed from the source video. Automatic captions, corrected for known mis-transcriptions.

Key terms

TPU
Tensor Processing Unit, a chip Google designs itself to run AI workloads.
inference
Running a trained AI model to produce answers, as opposed to training the model in the first place.
latency
The delay between sending a request and receiving a response.

Educational summary of a public statement. Not investment advice.