Back to models
publish

Phi-3

phi-3

Avg speed
18.27t/s
First token
0.53s
Total tests
5
Providers
1
Variants
1

Variants

Showing 1-1 of 1 providers

VariantSpeedLatencyTests
integrate.api.nvidia.com
18.27 t/s
0.53s
5

Recent test records

5 records
TimeModelSpeedLatency
Aug 13, 11:54 AMmicrosoft/phi-3-medium-128k-instruct
21.69 t/s
1.02s
Aug 13, 11:54 AMmicrosoft/phi-3-medium-128k-instruct
17.03 t/s
0.41s
Aug 13, 11:54 AMmicrosoft/phi-3-medium-128k-instruct
18.42 t/s
0.39s
Aug 13, 11:54 AMmicrosoft/phi-3-medium-128k-instruct
17.35 t/s
0.41s
Aug 13, 11:54 AMmicrosoft/phi-3-medium-128k-instruct
16.86 t/s
0.42s
microsoft/phi-3-medium-128k-instruct