⚡ WebGPU Benchmark Results (128.85x speedup) | Snowflake/snowflake-arctic-embed-s (fp16)
#97
by
Xenova
HF staff
- opened
| Batch Size | WASM (fp16) | WebGPU (fp16) |
| 1 | 2327.30 | 80.50 |
| 2 | 4652.50 | 244.20 |
| 4 | 9406.50 | 275.70 |
| 8 | 18884.60 | 369.10 |
| 16 | 37666.60 | 373.40 |
| 32 | 74977.30 | 581.90 |
- 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=