⚡ WebGPU Benchmark Results (1.76x speedup) – M1 Max WebGPU, Xenova/gte-base
#62
by
pcuenq
HF staff
- opened
| Batch Size | WebGPU (fp16) | WebGPU (fp32) |
| 1 | 38.70 | 61.20 |
| 2 | 69.70 | 120.20 |
| 4 | 133.10 | 220.90 |
| 8 | 254.30 | 426.50 |
| 16 | 491.80 | 830.50 |
| 32 | 949.30 | 1621.40 |
| 64 | 1926.20 | 3348.30 |
| 128 | 3787.00 | 6679.30 |
- Model: Xenova/gte-base
- Tests run: WebGPU (fp16), WebGPU (fp32)
- Sequence length: 512
- Browser: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36
- GPU: vendor=apple, architecture=common-3, device=, description=