修复awq 的mtp中的blockint8的问题

0cc9bc74 · yangql · 1b98d0bb · 0cc9bc74
Commit 0cc9bc74 authored Oct 15, 2025 by yangql
Show whitespace changes
Inline Side-by-side

Showing with 1 addition and 0 deletions

vllm/model_executor/layers/quantization/blockwise_int8.py vllm/model_executor/layers/quantization/blockwise_int8.py +1 -0

No files found.
--- a/vllm/model_executor/layers/quantization/blockwise_int8.py
+++ b/vllm/model_executor/layers/quantization/blockwise_int8.py
@@ -517,4 +517,5 @@ class BlockInt8MoEMethod:
            block_shape=self.quant_config.weight_block_size,
            use_nn_moe=use_nn_moe,
            shared_output=shared_output,
+            routed_scaling_factor=routed_scaling_factor,
        )
\ No newline at end of file