Support weight-only quantization with quantized operators in intel-extension-for-transformers. #533
Job | Run time |
---|---|
2m 46s | |
2m 45s | |
39s | |
35s | |
5m 30s | |
10m 29s | |
35s | |
41s | |
24m 0s |
Job | Run time |
---|---|
2m 46s | |
2m 45s | |
39s | |
35s | |
5m 30s | |
10m 29s | |
35s | |
41s | |
24m 0s |