LLM Inference on WebLLM Windows 11 RTX PRO 2000 Blackwell 8GB
51.1Decode Throughput (tok/s)Qwen2.5-0.5B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen2.5-0.5BBrowser=Chrome 144, Backend=D3D12, Quantization=q4f162026.02 | 51.1 | 650 | |
| Qwen2.5-1.5BBrowser=Chrome 144, Backend=D3D12, Quantization=q4f162026.02 | 45.7 | 350 | |
| Qwen2.5-0.5BBrowser=Firefox 147, Backend=D3D12, Quantization=q4f162026.02 | 9.1 | 73 | |
| Qwen2.5-1.5BBrowser=Firefox 147, Backend=D3D12, Quantization=q4f162026.02 | 9.1 | 55 |