After quantization, the model can inference on specific GPU(T4 V100 A100) but slower than without quantization, right?

[sisrfeng]

Quantization is currently only supported for CPUs.

Someone on zhihu says: Q: pytorch量化后的模型 能使用GPU进行预测吗？ A: 可以的，里面都是伪量化操作。 得到的预测结果应该还是对的，只是精度低了一点点， 会更慢 部署到FPGA/ASIC等特定硬件上才能看出加速效果