WebGPU Embedding Benchmark

Measure execution times of BERT models using WebGPU and WASM

What is WebGPU Embedding Benchmark ?

WebGPU Embedding Benchmark is a tool designed to measure the execution times of BERT models using WebGPU and WebAssembly (WASM). It helps developers and researchers evaluate the performance of embedding models in web-based environments, leveraging modern graphics technologies for accelerated computations.

Features

โ€ข WebGPU Acceleration: Leverages WebGPU for hardware-accelerated computations. โ€ข WASM Execution: Utilizes WebAssembly for efficient model inference. โ€ข Detailed Timing Measurements: Provides precise execution time metrics for model inference. โ€ข Cross-Platform Compatibility: Runs on modern web browsers supporting WebGPU. โ€ข Model Optimization Insights: Offers benchmarks to guide model optimization strategies. โ€ข Performance Comparison: Enables comparison of performance across different hardware setups.

How to use WebGPU Embedding Benchmark ?

  1. Set Up Environment: Ensure you have a modern web browser supporting WebGPU.
  2. Clone Repository: Clone the benchmark repository from its official source.
  3. Install Dependencies: Install required dependencies using npm or yarn.
  4. Run Benchmark: Execute the benchmark script to measure model performance.
  5. Analyze Results: Review the generated performance metrics and compare across different configurations.

Frequently Asked Questions

What does WebGPU Embedding Benchmark measure?
It measures the execution time of BERT models using WebGPU and WASM, providing insights into performance bottlenecks.

Which browsers support WebGPU?
As of now, browsers like Chrome, Firefox, and Edge provide experimental or full support for WebGPU.

Why is WebGPU combined with WASM for this benchmark?
WebGPU offers hardware acceleration, while WASM provides efficient computation, making them a powerful combination for high-performance web-based model inference.