diff --git a/lmdeploy/pytorch/backends/dlinfer/ascend/op_backend.py b/lmdeploy/pytorch/backends/dlinfer/ascend/op_backend.py
index e48ab79f18..2b1bba1464 100644
--- a/lmdeploy/pytorch/backends/dlinfer/ascend/op_backend.py
+++ b/lmdeploy/pytorch/backends/dlinfer/ascend/op_backend.py
@@ -158,7 +158,8 @@ def get_cpu_seqlens(is_decoding, is_unpaged_prefill):
                     repeat_interleave, used for attention metadata.
             """
             if is_decoding:
-                q_seqlens_cpu = None
+                q_seqlens_cpu = None if AscendOpsBackend.enable_graph else torch.arange(1, step_context.kv_seqlens.size(0) + 1,
+                                                                                           dtype=torch.int32)
                 kv_seqlens_cpu = kv_seqlens_expanded = step_context.kv_seqlens.cpu()
             elif is_unpaged_prefill:
                 q_seqlens_cpu = step_context.q_seqlens.cpu()
diff --git a/lmdeploy/pytorch/kernels/dlinfer/pagedattention.py b/lmdeploy/pytorch/kernels/dlinfer/pagedattention.py
index 13f4e12a58..4531b474ca 100644
--- a/lmdeploy/pytorch/kernels/dlinfer/pagedattention.py
+++ b/lmdeploy/pytorch/kernels/dlinfer/pagedattention.py
@@ -78,6 +78,7 @@ def paged_token_attention(
     k_cache,
     v_cache,
     attn_output,
+    q_seqlens,
     kv_seq_len,
     max_kv_seq_len,
     block_offsets,
@@ -96,6 +97,7 @@ def paged_token_attention(
         v_cache,
         block_offsets,
         block_size,
+        q_seqlens,
         kv_seq_len,
         max_kv_seq_len,
         num_q_heads,
@@ -167,6 +169,7 @@ def paged_attention_fwd(
             key_cache,
             value_cache,
             attn_output,
+            q_seqlens,
             kv_seqlens,
             max_kv_seq_len,
             block_offsets,
diff --git a/requirements/runtime_ascend.txt b/requirements/runtime_ascend.txt
index d94a38d0bf..22d1ca8418 100644
--- a/requirements/runtime_ascend.txt
+++ b/requirements/runtime_ascend.txt
@@ -22,9 +22,9 @@ safetensors
 sentencepiece
 shortuuid
 tiktoken
-torch>=2.3.1,<2.9.0
-torch-npu>=2.3.1,<2.9.0
-torchvision>=0.18.1,<0.24.0
+torch>=2.3.1,<2.10.0
+torch-npu>=2.3.1,<2.10.0
+torchvision>=0.18.1,<0.25.0
 transformers
 uvicorn
 xgrammar