From 087751a8f2abae0136ec351b86e2db5afd8ac149 Mon Sep 17 00:00:00 2001 From: Stefan He Date: Thu, 8 May 2025 14:45:33 -0700 Subject: [PATCH] Remove unecessary is_fa3_supported check (#6112) --- sgl-kernel/python/sgl_kernel/flash_attn.py | 4 ---- 1 file changed, 4 deletions(-) diff --git a/sgl-kernel/python/sgl_kernel/flash_attn.py b/sgl-kernel/python/sgl_kernel/flash_attn.py index e849d0df0..5a443ab43 100644 --- a/sgl-kernel/python/sgl_kernel/flash_attn.py +++ b/sgl-kernel/python/sgl_kernel/flash_attn.py @@ -144,10 +144,6 @@ def flash_attn_with_kvcache( logsumexp of each row of the matrix QK^T * scaling (e.g., log of the softmax normalization factor). """ - if not is_fa3_supported(): - raise NotImplementedError( - "flash_attn at sgl-kernel is only supported on sm90 and cu123 above" - ) assert k_cache.stride(-1) == 1, "k_cache must have contiguous last dimension" assert v_cache.stride(-1) == 1, "v_cache must have contiguous last dimension" if softmax_scale is None: