⚡ WebGPU Benchmark Results (119.64x speedup) | Snowflake/snowflake-arctic-embed-s (fp16)
#96
by
Xenova
HF staff
- opened
| Batch Size | WASM (fp16) | WebGPU (fp16) |
| 1 | 2337.30 | 134.50 |
| 2 | 4624.00 | 201.80 |
| 4 | 9333.80 | 340.70 |
| 8 | 18622.10 | 352.00 |
| 16 | 37052.30 | 386.40 |
| 32 | 74009.30 | 618.60 |
- Model: Snowflake/snowflake-arctic-embed-s
- 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/123.0.0.0 Safari/537.36
- GPU: vendor=nvidia, architecture=turing, device=, description=