Skip to content

Commit 5b5a724

Browse files
committed
minor fixing on tuning constraints
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
1 parent acd8694 commit 5b5a724

File tree

1 file changed

+1
-0
lines changed

1 file changed

+1
-0
lines changed

tensorrt_llm/_torch/custom_ops/torch_custom_ops.py

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -44,6 +44,7 @@ class MoERunner(TunableRunner):
4444
dynamic_tensor_specs=(DynamicTensorSpec(
4545
0, 0, get_last_power_of_2_num_tokens_buckets,
4646
last_positive_power_of_2), ),
47+
constraint_specs=(ConstraintSpec(1, 0, lambda shapes: shapes[0][0]), ),
4748
tune_max_num_tokens=8192,
4849
inputs_pre_hook=inputs_pre_hook,
4950
)

0 commit comments

Comments
 (0)