mirror of https://github.com/LTTLabsOSS/markbench-tests.git synced 2026-01-06 20:53:52 -05:00

Files

j-lin-lmg b229c4f027 Ffmpeg cpu benchmark (#179 )

Added ffmpeg cpu benchmark based on @nharris-lmg commands

3 options for encode (h264, av1, h265)

will report final vmaf score as "score" in report json

logs saved to artifacts

2025-12-23 16:20:51 -08:00

config

updated text generation benchmark to work for both openvino and windows ML, openvino remains to be tested

2024-12-23 16:19:20 -08:00

manifest.yaml

updated the manifest files for new test types

2024-12-24 16:06:05 -08:00

README.md

James round seconds (#129 )

2025-04-09 14:34:19 -07:00

ulprocai_text_gen.py

Ffmpeg cpu benchmark (#179 )

2025-12-23 16:20:51 -08:00

utils.py

Procyon parameters (#146 )

2025-06-02 12:35:09 -07:00

README.md

UL Procyon AI Text Generation

Runs the UL Procyon AI Text Generation benchmark using a specified engine and reads the Performance Score result from the output.

Prerequisites

Python 3.10+
UL Procyon installed in default location and activated with at least the AI tests
AI Text Generation Benchmark DLC installed

Options

--engine Specifies the hardware to benchmark.

Output

report.json

start_time: number representing a timestamp of the test's start time in milliseconds
end_time: number representing a timestamp of the test's end time in milliseconds
test: The name of the selected benchmark
test_version: The version of the benchmark
procyon_version: The version of Procyon used
score: The text generation scores