⚡ WebGPU Benchmark Results (37.69x speedup) - jina-embeddings-v2-small-en
#64
by
Xenova
HF staff
- opened
| Batch Size | WASM (fp32) | WebGPU (fp32) |
| 1 | 696.40 | 25.40 |
| 2 | 1390.40 | 72.90 |
| 4 | 2807.40 | 303.30 |
| 8 | 5647.60 | 308.80 |
| 16 | 11579.60 | 496.30 |
| 32 | 23855.60 | 633.00 |
- Model: Xenova/jina-embeddings-v2-small-en
- 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/122.0.0.0 Safari/537.36
- GPU: vendor=nvidia, architecture=turing, device=, description=