ResearchTasksVLA InferenceVLA InferenceFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedVLA Inference Efficiency 224x224 RGB, 256 tokens, single H100 GPU baseline vs proposedBLURR-Pi-017.1Latency (ms)5Feb 26, 2026
VLA Inference Efficiency 224x224 RGB, 256 tokens, single H100 GPU baseline vs proposedBLURR-Pi-017.1Latency (ms)5Feb 26, 2026