Skip to content

[None][feature] AutoDeploy: tighter MoE UT thresholds #9195

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Nov 19, 2025
Merged

[None][feature] AutoDeploy: tighter MoE UT thresholds #9195

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
41 changes: 29 additions & 12 deletions tests/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_trtllm_moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -119,16 +119,16 @@ def relu2(x: torch.Tensor) -> torch.Tensor:


def _get_test_data(
otype, wtype, batch_size, hidden_size, num_experts, intermediate_size, X_GEN_SCALE
otype, wtype, batch_size, hidden_size, num_experts, intermediate_size, X_GEN_SCALE, W_GEN_SCALE
):
input_shape = (batch_size, hidden_size)
w31_shape = (num_experts, 2 * intermediate_size, hidden_size)
w2_shape = (num_experts, hidden_size, intermediate_size)

x = cast_to_representable(gen_tensor(input_shape, otype, scale=X_GEN_SCALE))
router_logits = gen_tensor((batch_size, num_experts), otype)
w31_weight = gen_tensor(w31_shape, otype, wtype)
w2_weight = gen_tensor(w2_shape, otype, wtype)
w31_weight = gen_tensor(w31_shape, otype, wtype, W_GEN_SCALE)
w2_weight = gen_tensor(w2_shape, otype, wtype, W_GEN_SCALE)
w31_empty_scales = torch.empty(num_experts, 2, dtype=otype).cuda()
w2_empty_scales = torch.empty(num_experts, 1, dtype=otype).cuda()
return x, router_logits, w31_weight, w2_weight, w31_empty_scales, w2_empty_scales
Expand Down Expand Up @@ -203,9 +203,17 @@ def test_trtllm_fused_moe(
X_GEN_SCALE = 1.0
else:
X_GEN_SCALE = 0.5
W_GEN_SCALE = 0.1

x, router_logits, w31_weight, w2_weight, w31_scales, w2_scales = _get_test_data(
otype, wtype, batch_size, hidden_size, num_experts, intermediate_size, X_GEN_SCALE
otype,
wtype,
batch_size,
hidden_size,
num_experts,
intermediate_size,
X_GEN_SCALE,
W_GEN_SCALE,
)

routing_weights, selected_experts = compute_routing(router_logits, top_k)
Expand Down Expand Up @@ -278,14 +286,14 @@ def get_fc1_expert_weights(
w1_weight.contiguous(),
w2_weight.contiguous(),
)[0].view(x.shape)
torch.testing.assert_close(output_triton_moe, ad_test_output, rtol=1e-1, atol=1e-1)
torch.testing.assert_close(output_triton_moe, ad_test_output, rtol=1e-2, atol=1e-2)

diff = (ref_output - ad_test_output).abs()
print(f"max diff: {diff.max()}")
torch.testing.assert_close(ad_test_output, trtllm_test_output, rtol=1e-6, atol=1e-6)

_print_diff_if(lambda diff: diff.max() > 1e-1, diff, ad_test_output, ref_output)
torch.testing.assert_close(ref_output, ad_test_output, rtol=1e-1, atol=1e-1)
torch.testing.assert_close(ref_output, ad_test_output, rtol=1e-2, atol=1e-2)


FP8_TEST_DTYPES = [
Expand All @@ -305,7 +313,7 @@ def get_fc1_expert_weights(
not fp8_compatible() or not trtllm_ops_available(),
reason="Requires fp8 and trtllm support",
)
def test_trtllm_fused_fp8moe(
def test_trtllm_fused_moe_fp8(
batch_size,
hidden_size,
num_experts,
Expand Down Expand Up @@ -333,16 +341,18 @@ def test_trtllm_fused_fp8moe(
else:
X_GEN_SCALE = 0.5

def dequantize_weights(w31_weight, w2_weight, w31_scales, w2_scales):
W_GEN_SCALE = 0.1

def dequantize_weights(w31_weight, w2_weight, w31_scales, w2_scales, W_GEN_SCALE):
# input_shape = (batch_size, hidden_size)
w31_shape = (num_experts, 2 * intermediate_size, hidden_size)
w2_shape = (num_experts, hidden_size, intermediate_size)

w31_dequantized = gen_tensor(w31_weight.shape, otype)
w2_dequantized = gen_tensor(w2_weight.shape, otype)
for expert_id in range(num_experts):
w31 = cast_to_representable(gen_tensor(w31_shape[1:], otype, scale=0.1))
w2 = cast_to_representable(gen_tensor(w2_shape[1:], otype, scale=0.09))
w31 = cast_to_representable(gen_tensor(w31_shape[1:], otype, scale=W_GEN_SCALE))
w2 = cast_to_representable(gen_tensor(w2_shape[1:], otype, scale=W_GEN_SCALE))
w31_quant, s31 = dynamic_per_tensor_fp8_quant(w31)
w2_quant, s2 = dynamic_per_tensor_fp8_quant(w2)
w31_weight.data[expert_id].copy_(w31_quant)
Expand All @@ -354,11 +364,18 @@ def dequantize_weights(w31_weight, w2_weight, w31_scales, w2_scales):
return w31_dequantized, w2_dequantized

x, router_logits, w31_weight, w2_weight, w31_scales, w2_scales = _get_test_data(
otype, wtype, batch_size, hidden_size, num_experts, intermediate_size, X_GEN_SCALE
otype,
wtype,
batch_size,
hidden_size,
num_experts,
intermediate_size,
X_GEN_SCALE,
W_GEN_SCALE,
)

w31_dequantized, w2_dequantized = dequantize_weights(
w31_weight, w2_weight, w31_scales, w2_scales
w31_weight, w2_weight, w31_scales, w2_scales, W_GEN_SCALE
)

routing_weights, selected_experts = compute_routing(router_logits, top_k)
Expand Down