⚡ WebGPU Benchmark Results (29.90x speedup)
#42
by
a414166402
- opened
Batch Size | WASM (ms) | WebGPU (ms) |
1 | 457.30 | 21.60 |
2 | 933.00 | 81.10 |
4 | 1865.90 | 94.60 |
8 | 3791.30 | 205.60 |
16 | 7975.80 | 345.60 |
32 | 16339.70 | 546.40 |
- Model: Xenova/all-MiniLM-L6-v2
- Quantized: false
- Sequence length: 512
- Browser: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36
- GPU: vendor=nvidia, architecture=turing, device=, description=