⚡ WebGPU Benchmark Results (70.35x speedup)
#90
by
Xenova
HF staff
- opened
Batch Size | WASM (fp32) | WebGPU (fp32) |
1 | 1068.30 | 115.60 |
2 | 2132.10 | 261.40 |
4 | 4272.90 | 276.20 |
8 | 8479.80 | 291.50 |
16 | 16798.50 | 380.20 |
32 | 33246.30 | 472.60 |
- Model: Snowflake/snowflake-arctic-embed-xs
- Tests run: WASM (fp32), WebGPU (fp32)
- Sequence length: 512
- Browser: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36
- GPU: vendor=nvidia, architecture=turing, device=, description=