Sail Research
@sailresearchco
Infra for long-horizon agents.
87 Following 4.1K Followers
We nearly doubled MFU on TPU v6e serving Gemma 4 31B, taking prefill from 32% to 63% MFU, and 2x throughput.
v6e has 2.5x less HBM and less than half the memory bandwidth of an H100.
@transmissions11 wrote it up here:
Show more
