diff --git a/qwen3_6_scripts/patch_xformers_sdpa_seq.py b/qwen3_6_scripts/patch_xformers_sdpa_seq.py
index 8de1565a..af7037c2 100644
--- a/qwen3_6_scripts/patch_xformers_sdpa_seq.py
+++ b/qwen3_6_scripts/patch_xformers_sdpa_seq.py
@@ -200,6 +200,10 @@ FALLBACK_METHOD = '''
         max_seqlen = max(seq_lens_list)
 
         try:
+            # Skip flash_attn during profiling — OOMs on large dummy batch
+            import os
+            if os.environ.get("BI100_IN_STARTUP_PROFILE") == "1":
+                raise RuntimeError("skip flash_attn during profiling")
             out = _ixf.flash_attn_varlen_func(
                 q_flat.to(torch.float16),
                 k_flat.to(torch.float16),