Step 2: Model Profiling

Step 2-1: Upload your AI model

Step 2-2: Select benchmark configurations

Select the desired device and click the Benchmark button.

🚧

Benchmarking may take some time.
You can safely leave the page after starting a benchmark. Once it’s complete, the benchmark status will be displayed at the bottom of the main page.

Step 2-3: Check benchmark result

Once the benchmark is complete, you can view a detailed performance summary like the example below.
It includes key metrics such as:

• Inference latency (avg): Time per inference in milliseconds
• Memory footprint: Peak memory used during inference
• File size of the optimized model
• Device info: Including runtime environment, CPU/GPU/NPU specs, and memory