biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[CI Failure] Fix test_flashinfer_cutlass_mxfp4_mxfp8_fused_moe (#24750)

Browse Source 
Signed-off-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
Michael Goin
2025-09-13 03:29:19 -04:00
committed by GitHub
parent 1da0f1441d
commit 59d7ffc17f
2 changed files with 3 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/kernels/moe/test_mxfp4_moe.py
View File

@@ -771,11 +771,11 @@ def test_flashinfer_cutlass_mxfp4_mxfp8_fused_moe(
w13_ref = dequant_mxfp4_batches(
w13_q.view(torch.uint8),
w13_scale.view(torch.uint8).reshape(-1)).to(torch.float32).reshape(
num_experts, 2 * intermediate_size, hidden_size)
num_experts, 2 * intermediate_size, hidden_size).to(device)
w2_ref = dequant_mxfp4_batches(
w2_q.view(torch.uint8),
w2_scale.view(torch.uint8).reshape(-1)).to(torch.float32).reshape(
num_experts, hidden_size, intermediate_size)
num_experts, hidden_size, intermediate_size).to(device)
# Quantize activations for SM100 path and dequantize for reference
hidden_states_q, hidden_states_sf = mxfp8_quantize(hidden_states, True, 32)