⚡ WebGPU Benchmark Results (48.45x speedup) - jina-embeddings-v2-small-en (fp16)
#66
by
Xenova
- opened
| Batch Size | WASM (fp16) | WebGPU (fp16) |
| 1 | 1640.10 | 83.10 |
| 2 | 3223.60 | 224.80 |
| 4 | 6432.30 | 416.10 |
| 8 | 12957.20 | 454.40 |
| 16 | 25993.90 | 796.00 |
| 32 | 51885.80 | 1070.90 |
- Model: Xenova/jina-embeddings-v2-small-en
- Tests run: WASM (fp16), WebGPU (fp16)
- Sequence length: 512
- Browser: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36
- GPU: vendor=nvidia, architecture=turing, device=, description=