diff --git a/unsloth/models/llama.py b/unsloth/models/llama.py
index 4d8ec1367e..f19609fa44 100644
--- a/unsloth/models/llama.py
+++ b/unsloth/models/llama.py
@@ -247,7 +247,7 @@ def LlamaAttention_fast_forward_inference(
 
     # Grouped query attention
     _, _, cached_len, _ = Knn.shape
-    if bsz == 1 or not SDPA_HAS_GQA and n_groups != 1:
+    if True:
         Knn = Knn[:, :, None, :, :].expand(bsz, n_kv_heads, n_groups, cached_len, head_dim)
         Vnn = Vnn[:, :, None, :, :].expand(bsz, n_kv_heads, n_groups, cached_len, head_dim)
         Knn = Knn.reshape(bsz, n_heads, cached_len, head_dim)
@@ -266,10 +266,7 @@ def LlamaAttention_fast_forward_inference(
         A[:] = torch_nn_functional_softmax(A, dim = -1, dtype = torch.float32)#.to(A.dtype)
         A = torch_matmul(A, Vnn, out = Qn)
     else:
-        if SDPA_HAS_GQA:
-            A = scaled_dot_product_attention(Qn, Knn, Vnn, attn_mask = attention_mask, is_causal = False, enable_gqa = True)
-        else:
-            A = scaled_dot_product_attention(Qn, Knn, Vnn, attn_mask = attention_mask, is_causal = False)
+        A = scaled_dot_product_attention(Qn, Knn, Vnn, attn_mask = attention_mask, is_causal = False)
     pass
     A = A.transpose(1, 2)
     A = A.reshape(bsz, 1, attention_size)