how to use benchmark_app test qwen2.5-7B-instruct model inference performance?

[Light-Travlling]

i have deployed "qwen2.5-7B-instruct" model on my openvino environment，and successfully！Now i want to test the inference performance，can i use “benchmark_app” to do it?
Using command "benchmark_app -m ./Qwen2.5-7B-Instruct-fp16-ov/openvino_model.xml -hint latency", it wrong like following picture. i need help

[ilya-lavrenov]

Please, use a dedicated benchmark for LLMs from GenAI repo
https://github.com/openvinotoolkit/openvino.genai/blob/master/samples/cpp/text_generation/benchmark_genai.cpp

[Light-Travlling]

thank you very much, however i don't find mutil requset result, it seems single batch. Where to find i want

[ilya-lavrenov]

For multi request, you need to use continuous matching to maximize tput https://github.com/openvinotoolkit/openvino.genai/tree/master/tools/continuous_batching/benchmark

[Light-Travlling]

thank you very much