Skip to content

Food & Beverage QC

Render sharpness and perceived quality for menu and delivery-app imagery. Benchmark set not yet measured.

Benchmark last measured 1 day ago · pipeline v3.0 · re-measured manually on every model or pipeline change.

Objective metrics

computed

Each figure is computed by a named library against a stated threshold. Hard presets first — their lower numbers are the point, not a thing to hide. An unmeasured cell shows a placeholder, never an invented number.

Menu Hero Shot

oaktree/image-edit · $0.04/image

awaiting benchmark

No food benchmark set exists yet — every figure below is an honest placeholder until the food vertical gets its own measured run.

Render sharpness (BRISQUE)

target threshold pending calibration

PLACEHOLDER · run harness n=0 · date TBD

Perceived quality (CLIP-IQA)

target threshold pending calibration

PLACEHOLDER · run harness n=0 · date TBD

Auto-QC first-pass rate

target ≥ 85% first pass

PLACEHOLDER · run harness n=0 · date TBD

No benchmark renders exist for this vertical yet. The scorecards above stay as placeholders until a measured run lands — we don't publish numbers we haven't measured.

Theme
Type scale
A A