Skip to content

[ModelOpt] Remove NVFP4 MoE K%16==0 constraint #26891

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
Oct 15, 2025
Merged

[ModelOpt] Remove NVFP4 MoE K%16==0 constraint #26891

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
12 changes: 0 additions & 12 deletions vllm/model_executor/layers/quantization/modelopt.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1542,23 +1542,11 @@ def process_weights_after_loading(self, layer: torch.nn.Module) -> None:
del layer.w2_input_scale_quant
else:
# Non-TRT-LLM processing (Cutlass or non-flashinfer)
assert layer.w13_weight_scale.shape[2] % 16 == 0, (
"Expected weight_scale.dim(1) to be divisible by 16"
)
assert layer.w13_weight_scale.dtype == torch.float8_e4m3fn, (
"Weight Blockscale must be represented as FP8-E4M3"
)
w13_blockscale_swizzled = swizzle_blockscale(layer.w13_weight_scale)
layer.w13_weight_scale = Parameter(
w13_blockscale_swizzled, requires_grad=False
)

assert layer.w2_weight_scale.shape[2] % 16 == 0, (
"Expected weight_scale.dim(1) to be divisible by 16"
)
assert layer.w2_weight_scale.dtype == torch.float8_e4m3fn, (
"Weight Blockscale must be represented as FP8-E4M3"
)
w2_blockscale_swizzled = swizzle_blockscale(layer.w2_weight_scale)
layer.w2_weight_scale = Parameter(
w2_blockscale_swizzled, requires_grad=False
Expand Down