Skip to navigation

Performance Results

View as MarkdownOpen in Claude

Performance data

The following table shows the performance data for the LipSync NIM for concurrent streams on a single GPU with the provided sample input files and the Python client. The LipSync NIM supports concurrent streams per GPU. The number of concurrent streams can be configured using the NIM_MAX_CONCURRENCY_PER_GPU environment variable (default: 1). Higher concurrency values consume more GPU memory and can cause out-of-memory errors.

GPUInput Video ResolutionConcurrencyAverage FPS
RTX PRO 6000 Blackwell Server Edition720p163
720p237
720p419
1080p166
1080p238
1080p419
4K166
4K235
L40S720p161
720p234
720p417
1080p162
1080p233
1080p417
4K160
4K231
A40720p132
720p217
720p49
1080p133
1080p217
1080p49
4K131
4K216
L4720p125
720p213
720p47
1080p125
1080p213
1080p47
4K125
4K212
A10G720p124
720p213
720p46
1080p124
1080p213
1080p46
4K124
4K212
RTX 5090720p174
720p243
720p422
1080p177
1080p242
1080p421
4K175
4K241
RTX 5080720p160
720p234
720p417
1080p161
1080p234
1080p417
4K160
4K232
RTX 4090720p165
720p237
720p419
1080p166
1080p237
1080p419
4K165
4K235

The inference FPS is calculated by dividing the total number of frames in input video file by the total inference time in seconds (measured from the time the request is sent until complete output file is received by the client).

Note: Video extension operations significantly increase processing time and memory usage due to frame buffering.

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