⚡ WebGPU Benchmark Results (72.52x speedup) | Snowflake/snowflake-arctic-embed-xs (fp16)
#95
by
Xenova
HF staff
- opened
| Batch Size | WASM (fp16) | WebGPU (fp16) |
| 1 | 1163.90 | 66.40 |
| 2 | 2313.90 | 192.80 |
| 4 | 4658.40 | 246.60 |
| 8 | 9289.00 | 240.80 |
| 16 | 18593.70 | 267.10 |
| 32 | 37226.10 | 513.30 |
- Model: Snowflake/snowflake-arctic-embed-xs
- 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=
Xenova
changed discussion title from
⚡ WebGPU Benchmark Results (72.52x speedup)
to ⚡ WebGPU Benchmark Results (72.52x speedup) | Snowflake/snowflake-arctic-embed-xs (fp16)