Paragon AI Live Demo

Experience type-generic neural networks with WebGPU acceleration and distributed growth

Loading Paragon WASM Engine...

WebGPU Acceleration

GPU vs CPU benchmarking

Type-Generic

Float32, Int32, Uint32 networks

Network Growth

Self-evolving architectures

ADHD Evaluation

Advanced performance metrics

1 Performance Benchmarking

Test CPU vs GPU performance across different network types and activation functions

Benchmark Settings

Single Forward Pass Timing

Each test runs one forward pass for precise GPU vs CPU measurement (like the working index.html demo)

Performance Results

Run a benchmark to see CPU vs GPU performance comparisons

About the Benchmark:

  • Tests single forward pass performance on different activation functions
  • Compares CPU vs WebGPU acceleration performance
  • Evaluates type-generic network efficiency (Float32, Int32, Uint32)
  • Results show milliseconds per forward pass and speedup ratios

2 GPU Stress Test

Run 10000 consecutive inputs through the GPU to test sustained performance under load

Stress Test Settings

High Load Performance Test

This test runs 10000 forward passes to stress test GPU performance and measure sustained throughput

Number of consecutive forward passes to run

Stress Test Results

Run the stress test to see sustained GPU performance metrics

Performance chart will appear here after running the stress test

šŸ”„ About the Stress Test:

  • Runs 10000+ consecutive forward passes to test GPU endurance
  • Measures sustained throughput and identifies performance degradation
  • Tests GPU memory management under continuous load
  • Compares first vs last 100 passes to detect thermal throttling
  • Shows real-time performance graphs and statistics