Skip to content

Enabled configurable auto Tensor Parallelism (TP) for the inference of diverse models #11491

Enabled configurable auto Tensor Parallelism (TP) for the inference of diverse models

Enabled configurable auto Tensor Parallelism (TP) for the inference of diverse models #11491

unit-tests (3.10)

succeeded Feb 5, 2025 in 1m 53s