diff --git a/python/sglang/srt/mem_cache/allocator.py b/python/sglang/srt/mem_cache/allocator.py
index ef102f3abb43..1eaa622219f2 100644
--- a/python/sglang/srt/mem_cache/allocator.py
+++ b/python/sglang/srt/mem_cache/allocator.py
@@ -307,10 +307,16 @@ def free(self, free_index: torch.Tensor):
             self.free_swa(free_index)
         else:
             self.free_group.append(free_index)
-        assert (
-            self.full_attn_allocator.available_size() <= self.full_attn_allocator.size
-        )
-        assert self.swa_attn_allocator.available_size() <= self.swa_attn_allocator.size
+
+        def _is_elastic(self, alloc):
+            return hasattr(alloc, "kvcached_allocator")
+
+        if not self._is_elastic(self.full_attn_allocator):
+            assert (
+                self.full_attn_allocator.available_size() <= self.full_attn_allocator.size
+            )
+        if not self._is_elastic(self.swa_attn_allocator):
+            assert self.swa_attn_allocator.available_size() <= self.swa_attn_allocator.size
 
     def free_swa(self, free_index: torch.Tensor):
         swa_indices = self.full_to_swa_index_mapping[free_index]