Enable Nvidia's ModelOpt fp8 quantized models #3682
merrymercypr-test.yml
on: pull_request
Matrix: unit-test-backend-1-gpu
unit-test-frontend
1m 35s
unit-test-backend-2-gpu
11m 27s
performance-test-1-gpu-part-1
1m 24s
performance-test-1-gpu-part-2
11m 16s
performance-test-2-gpu
9m 33s
accuracy-test-1-gpu
11m 27s
accuracy-test-2-gpu
6m 21s
finish
0s
Annotations
14 errors
|
performance-test-1-gpu-part-1
Process completed with exit code 1.
|
|
unit-test-frontend
Process completed with exit code 1.
|
|
unit-test-backend-1-gpu (0-6)
Process completed with exit code 1.
|
|
unit-test-backend-1-gpu (23-30)
The job was canceled because "_0-6" failed.
|
|
unit-test-backend-1-gpu (23-30)
The operation was canceled.
|
|
unit-test-backend-1-gpu (16-23)
The job was canceled because "_0-6" failed.
|
|
unit-test-backend-1-gpu (16-23)
The operation was canceled.
|
|
unit-test-backend-1-gpu (6-16)
The job was canceled because "_0-6" failed.
|
|
unit-test-backend-1-gpu (6-16)
The operation was canceled.
|
|
unit-test-backend-1-gpu (30-100)
The job was canceled because "_0-6" failed.
|
|
unit-test-backend-1-gpu (30-100)
The operation was canceled.
|
|
performance-test-1-gpu-part-2
The action 'Benchmark offline throughput (w/o RadixAttention)' has timed out after 10 minutes.
|
|
unit-test-backend-2-gpu
The action 'Evaluate data parallelism accuracy (DP=2)' has timed out after 10 minutes.
|
|
accuracy-test-1-gpu
Process completed with exit code 1.
|