Matthew Kotila

Unfortunately we don't support supplying binary files for more than one request, but you should be able to convert the binary data into b64 representation and include that in an...

> @kzelias: @matthewkotila > If I use b64 + json, I will need to change the logic of the triton service, right? Would need to decode b64. The decoding of...

@siretru you can find information about the GPU utilization metric that Perf Analyzer offers here: https://github.com/triton-inference-server/client/blob/main/src/c%2B%2B/perf_analyzer/docs/measurements_metrics.md#server-side-prometheus-metrics