vLLM Inference Performance on Qwen3-0.6B
0.336Model Load Time (s)Safetensors
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Safetensors2025.12 | 0.336 | 90.512 | 197.15 | 10,557.58 | 1.17 | |
| CryptoTensorsEncryption=Unencrypted2025.12 | 0.34 | 91.1 | 195.3 | 10,558.89 | 1.17 | |
| CryptoTensorsEncryption=Encrypted2025.12 | 2.073 | 91.062 | 193.92 | 10,613.48 | 1.17 |