diff --git a/python/sglang/srt/layers/quantization/fp8_utils.py b/python/sglang/srt/layers/quantization/fp8_utils.py index 0bdecef0a..395549abe 100755 --- a/python/sglang/srt/layers/quantization/fp8_utils.py +++ b/python/sglang/srt/layers/quantization/fp8_utils.py @@ -68,6 +68,7 @@ def use_aiter_triton_gemm_w8a8_tuned_gfx950(n: int, k: int) -> bool: (4608, 7168), (512, 7168), (7168, 2048), + (7168, 2304), (7168, 16384), (7168, 256), (8192, 1024),