From 78043d444837a231a108f4049dafbc6c9e2545a6 Mon Sep 17 00:00:00 2001 From: Ziang Li Date: Sat, 11 Apr 2026 04:11:28 -0700 Subject: [PATCH] [Misc] [MXFP8] Drop sm100 mxfp8 warning (#21881) --- python/sglang/srt/configs/model_config.py | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/python/sglang/srt/configs/model_config.py b/python/sglang/srt/configs/model_config.py index e90ee3881..67cce0e63 100644 --- a/python/sglang/srt/configs/model_config.py +++ b/python/sglang/srt/configs/model_config.py @@ -1071,8 +1071,10 @@ class ModelConfig: f"supported in ROCm." ) if self.quantization not in optimized_quantization_methods: - # Don't warn for MXFP4 on SM100 since it has optimized kernels - if not (self.quantization == "mxfp4" and is_sm100_supported()): + # Don't warn for MXFP4/MXFP8 on SM100 since they have optimized kernels + if not ( + self.quantization in ["mxfp4", "mxfp8"] and is_sm100_supported() + ): logger.warning( "%s quantization is not fully " "optimized yet. The speed can be slower than "