vllm
a0e9ee83 - [Benchmark] Fix OOM during MoE kernel tuning for large models (#31604)

Commit

3 days ago

[Benchmark] Fix OOM during MoE kernel tuning for large models (#31604) Signed-off-by: Alfred <massif0601@gmail.com>

References

Author

massif-01

Parents