Back to models
publish

Phi-4

phi-4

Avg speed
53.44t/s
First token
1.03s
Total tests
15
Providers
2
Variants
2

Variants

Showing 1-2 of 2 providers

VariantSpeedLatencyTests
integrate.api.nvidia.com
74.19 t/s
0.46s
5
ai.api.xn--fiqs8s
43.07 t/s
1.32s
10

Recent test records

15 records
TimeModelSpeedLatency
Aug 13, 12:03 PMmicrosoft/phi-4-mini-flash-reasoning
74.62 t/s
0.94s
Aug 13, 12:03 PMmicrosoft/phi-4-mini-flash-reasoning
74.73 t/s
0.32s
Aug 13, 12:03 PMmicrosoft/phi-4-mini-flash-reasoning
72.66 t/s
0.35s
Aug 13, 12:03 PMmicrosoft/phi-4-mini-flash-reasoning
74.99 t/s
0.37s
Aug 13, 12:03 PMmicrosoft/phi-4-mini-flash-reasoning
73.94 t/s
0.32s
Feb 20, 07:45 AMPhi-4
41.81 t/s
1.23s
Feb 20, 07:45 AMPhi-4
41.66 t/s
0.84s
Feb 20, 07:45 AMPhi-4
41.59 t/s
0.84s
Feb 20, 07:45 AMPhi-4
42.24 t/s
0.84s
Feb 20, 07:45 AMPhi-4
40.79 t/s
0.74s
Feb 9, 04:17 AMPhi-4
44.55 t/s
3.32s
Feb 9, 04:17 AMPhi-4
44.30 t/s
0.80s
Feb 9, 04:17 AMPhi-4
44.55 t/s
0.91s
Feb 9, 04:17 AMPhi-4
44.94 t/s
2.87s
Feb 9, 04:17 AMPhi-4
44.31 t/s
0.81s
microsoft/phi-4-mini-flash-reasoning
Phi-4