Ornotto benchmark explorer

67 FontLab routing queries in 30 languages, five choices. Apple M4 Max, 48 GiB; hosted Jev is a remote reference. Scores and timings are single recorded passes, not confidence intervals. Weight GB is disk size, not peak RAM. Classification latency excludes translation and loading.

Scoring modeExecution deviceEngineModel familyMinimum correct / 67Maximum ms/querySearch runsHorizontal scalePareto frontier only

Loading interactive charts…

C = CPU · G = GPU · C+G = mixed · C+N = CPU + Neural Engine · A = automatic CPU/GPU/Neural Engine placement · R = remote · ? = CPU/GPU unrecorded. Configuration describes the run; it is not a hardware utilization trace.

Hover a point for its exact run. Click a point or a table row for execution evidence.

Filtered runs, fastest first. Pareto membership is calculated within the current filters.
Run	C/G	Correct / 67	ms/query	Weight GB	Pareto	Readout

Generated from the public benchmark JSON. Charts: Plotly.js 3.1.0 (MIT). Methodology and current report.