Skip to navigation

Performance Results

View as MarkdownOpen in Claude

The NVIDIA Eye Contact NIM supports multiple concurrent inputs per GPU. The following table shows the performance data for various concurrency levels on various GPUs with the provided sample input files and the Python client.

GPUInput Video ResolutionConcurrencyAverage FPS
NVIDIA RTX PRO 6000 Blackwell Server Edition720p166
720p260
720p442
720p820
1080p151
1080p242
1080p431
1080p815
RTX L40720p161
720p254
720p439
720p819
1080p142
1080p234
1080p426
1080p814
RTX L4720p160
720p251
720p451
720p826
1080p141
1080p236
1080p428
1080p813
RTX A40720p160
720p252
720p448
720p824
1080p138
1080p232
1080p422
1080p811
RTX A10720p158
720p253
720p439
720p818
1080p142
1080p235
1080p426
1080p813
RTX A2720p152
720p246
720p433
720p821
1080p135
1080p226
1080p420
1080p810
RTX T4720p140
720p233
720p430
720p818
1080p129
1080p221
1080p418
1080p811

The average FPS is calculated by dividing the sum of number of frames in the input video files by the total inference time in seconds (measured from the time the concurrent requests are sent until complete output files are received by the client). This metric helps determine the effective performance scaling when running multiple inputs simultaneously.

For more information, refer to the Maxine NIM clients Github repository: NVIDIA-Maxine/nim-clients.