diff --git a/benchmarks/yaml/qwen25_7b-vl-32k-bf16.yaml b/benchmarks/yaml/qwen25_7b-vl-32k-bf16.yaml
new file mode 100644
index 00000000000..d159e676f60
--- /dev/null
+++ b/benchmarks/yaml/qwen25_7b-vl-32k-bf16.yaml
@@ -0,0 +1,6 @@
+max_model_len: 32768
+max_num_seqs: 128
+gpu_memory_utilization: 0.85
+tensor_parallel_size: 1
+limit_mm_per_prompt: '{"image": 100, "video": 100}'
+enable_mm: True
\ No newline at end of file
diff --git a/benchmarks/yaml/request_yaml/qwen25-vl-32k.yaml b/benchmarks/yaml/request_yaml/qwen25-vl-32k.yaml
new file mode 100644
index 00000000000..0c9a944e699
--- /dev/null
+++ b/benchmarks/yaml/request_yaml/qwen25-vl-32k.yaml
@@ -0,0 +1,8 @@
+top_p: 0.8
+temperature: 0.7
+metadata:
+  min_tokens: 1
+max_tokens: 32768
+repetition_penalty: 1.05
+frequency_penalty: 0
+presence_penalty: 0
\ No newline at end of file