Support weight-only quantization with quantized operators in intel-extension-for-transformers. #534
Job | Run time |
---|---|
2m 56s | |
3m 23s | |
42s | |
41s | |
3m 13s | |
3m 16s | |
35s | |
43s | |
15m 29s |
Job | Run time |
---|---|
2m 56s | |
3m 23s | |
42s | |
41s | |
3m 13s | |
3m 16s | |
35s | |
43s | |
15m 29s |