A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
Add data_parallel_size to VllmConfig string representation (#24298)
Co-authored-by: Cong Chen <congc@meta.com>
cong-meta committed
Sep 8, 2025 at 21:35 UTC
b2f7745774fc3d5105c0e0babbfb656bfcacfcac
1 file changed
+1
vllm/config/__init__.py
+1
@@ -3790,6 +3790,7 @@ class VllmConfig:
3790
f"load_format={self.load_config.load_format}, "
3791
f"tensor_parallel_size={self.parallel_config.tensor_parallel_size}, " # noqa
3792
f"pipeline_parallel_size={self.parallel_config.pipeline_parallel_size}, " # noqa
3793
+ f"data_parallel_size={self.parallel_config.data_parallel_size}, " # noqa
3794
f"disable_custom_all_reduce={self.parallel_config.disable_custom_all_reduce}, " # noqa
3795
f"quantization={self.model_config.quantization}, "
3796
f"enforce_eager={self.model_config.enforce_eager}, "