diff --git a/tests/test_infer/test_config_and_struct.py b/tests/test_infer/test_config_and_struct.py
index e0736518c..47d3839e4 100755
--- a/tests/test_infer/test_config_and_struct.py
+++ b/tests/test_infer/test_config_and_struct.py
@@ -63,6 +63,9 @@ def check_config_and_inference():
     batch.add_seqs([sequence])
     batch.add_seqs([sequence2, sequence3])
 
+    # add duplicated sequence to test that it will not be counted twice
+    batch.add_seqs([sequence])
+
     assert batch.is_empty == False
     assert batch.get_batch_size() == 3
     batch.update_batch_tokens([1, 2, 3])
diff --git a/tests/test_infer_ops/triton/kernel_utils.py b/tests/test_infer/test_ops/triton/kernel_utils.py
similarity index 100%
rename from tests/test_infer_ops/triton/kernel_utils.py
rename to tests/test_infer/test_ops/triton/kernel_utils.py
diff --git a/tests/test_infer_ops/triton/test_context_attn_unpad.py b/tests/test_infer/test_ops/triton/test_context_attn_unpad.py
similarity index 98%
rename from tests/test_infer_ops/triton/test_context_attn_unpad.py
rename to tests/test_infer/test_ops/triton/test_context_attn_unpad.py
index 0a3ede555..b529e76d1 100644
--- a/tests/test_infer_ops/triton/test_context_attn_unpad.py
+++ b/tests/test_infer/test_ops/triton/test_context_attn_unpad.py
@@ -6,7 +6,7 @@ from transformers.modeling_attn_mask_utils import AttentionMaskConverter
 from colossalai.inference.modeling.layers.attention import PagedAttention
 from colossalai.kernel.triton import context_attention_unpadded
 from colossalai.utils import get_current_device
-from tests.test_infer_ops.triton.kernel_utils import generate_caches_and_block_tables_v2, torch_attn_ref
+from tests.test_infer.test_ops.triton.kernel_utils import generate_caches_and_block_tables_v2, torch_attn_ref
 
 try:
     import triton  # noqa
diff --git a/tests/test_infer_ops/triton/test_decoding_attn.py b/tests/test_infer/test_ops/triton/test_decoding_attn.py
similarity index 99%
rename from tests/test_infer_ops/triton/test_decoding_attn.py
rename to tests/test_infer/test_ops/triton/test_decoding_attn.py
index 5eac026bb..4b9b63f7d 100644
--- a/tests/test_infer_ops/triton/test_decoding_attn.py
+++ b/tests/test_infer/test_ops/triton/test_decoding_attn.py
@@ -4,7 +4,7 @@ from packaging import version
 
 from colossalai.kernel.triton import flash_decoding_attention
 from colossalai.utils import get_current_device
-from tests.test_infer_ops.triton.kernel_utils import (
+from tests.test_infer.test_ops.triton.kernel_utils import (
     convert_kv_unpad_to_padded,
     generate_caches_and_block_tables_v2,
     prepare_padding_mask,
diff --git a/tests/test_infer_ops/triton/test_fused_rotary_embedding.py b/tests/test_infer/test_ops/triton/test_fused_rotary_embedding.py
similarity index 100%
rename from tests/test_infer_ops/triton/test_fused_rotary_embedding.py
rename to tests/test_infer/test_ops/triton/test_fused_rotary_embedding.py
diff --git a/tests/test_infer_ops/triton/test_kvcache_copy.py b/tests/test_infer/test_ops/triton/test_kvcache_copy.py
similarity index 97%
rename from tests/test_infer_ops/triton/test_kvcache_copy.py
rename to tests/test_infer/test_ops/triton/test_kvcache_copy.py
index 3b0a0f765..5612f2bd9 100644
--- a/tests/test_infer_ops/triton/test_kvcache_copy.py
+++ b/tests/test_infer/test_ops/triton/test_kvcache_copy.py
@@ -5,7 +5,7 @@ from packaging import version
 from colossalai.inference.modeling.layers.attention import copy_to_cache
 from colossalai.kernel.triton import copy_kv_to_blocked_cache
 from colossalai.utils import get_current_device
-from tests.test_infer_ops.triton.kernel_utils import generate_caches_and_block_tables_v2, mock_alloc_single_token
+from tests.test_infer.test_ops.triton.kernel_utils import generate_caches_and_block_tables_v2, mock_alloc_single_token
 
 try:
     import triton  # noqa
diff --git a/tests/test_infer_ops/triton/test_rmsnorm_triton.py b/tests/test_infer/test_ops/triton/test_rmsnorm_triton.py
similarity index 100%
rename from tests/test_infer_ops/triton/test_rmsnorm_triton.py
rename to tests/test_infer/test_ops/triton/test_rmsnorm_triton.py
diff --git a/tests/test_infer_ops/triton/test_rotary_embdding_unpad.py b/tests/test_infer/test_ops/triton/test_rotary_embdding_unpad.py
similarity index 98%
rename from tests/test_infer_ops/triton/test_rotary_embdding_unpad.py
rename to tests/test_infer/test_ops/triton/test_rotary_embdding_unpad.py
index 529c9fb2f..6a8dc85f0 100644
--- a/tests/test_infer_ops/triton/test_rotary_embdding_unpad.py
+++ b/tests/test_infer/test_ops/triton/test_rotary_embdding_unpad.py
@@ -4,7 +4,7 @@ from packaging import version
 from transformers.models.llama.modeling_llama import LlamaRotaryEmbedding, apply_rotary_pos_emb
 
 from colossalai.kernel.triton import rotary_embedding
-from tests.test_infer_ops.triton.kernel_utils import mock_alloc_block_table_and_kvcache_v2
+from tests.test_infer.test_ops.triton.kernel_utils import mock_alloc_block_table_and_kvcache_v2
 
 try:
     import triton  # noqa
diff --git a/tests/test_infer_ops/triton/test_xine_copy.py b/tests/test_infer/test_ops/triton/test_xine_copy.py
similarity index 100%
rename from tests/test_infer_ops/triton/test_xine_copy.py
rename to tests/test_infer/test_ops/triton/test_xine_copy.py