---
name: APCA
slug: apca
category: color
status: published
created: 2026-08-21T00:00:00.000Z
modified: 2026-08-21T00:00:00.000Z
definition: A perceptual contrast algorithm proposed for WCAG 3 that scores a
  text and background pair as an Lc value and takes font size and weight into
  account.
aliases:
  - name: Advanced Perceptual Contrast Algorithm
  - name: Lc value
    source: community
  - name: WCAG 3 contrast
    source: community
  - name: perceptual contrast
tags:
  - a11y
  - perception
  - wcag
relations:
  contrastWith:
    - color-vision-deficiency
    - conformance-level
    - large-text-threshold
    - contrast-ratio
  variantOf: []
  partOf: []
  seeAlso: []
implementations:
  - system: radix
    name: Lc 60 and Lc 90 guarantee
    url: https://www.radix-ui.com/colors/docs/palette-composition/understanding-the-scale
sources:
  - title: "Radix Colors: understanding the scale"
    url: https://www.radix-ui.com/colors/docs/palette-composition/understanding-the-scale
demo: inline
exhibit: false
useWhen: scoring contrast in a way that accounts for text size and weight
---

APCA reports one number, Lc, on a scale running to about 106 for black on white and about
minus 108 for white on black. The sign is the point: the algorithm knows which of the pair
is the text, so light text on a dark field and dark text on a light field are two different
measurements rather than one. The [contrast ratio](/contrast-ratio) in WCAG 2 is symmetric
by construction, dividing the lighter [relative luminance](/relative-luminance) by the
darker, and cannot tell those two situations apart at all.

The second departure is that a score alone is not a verdict. APCA pairs its output with a
lookup table of font sizes and weights, so the same pair of colours can be fine for a 24px
bold heading and much too weak for 14px body text. Roughly, thin small text is asked for
around Lc 90, ordinary body text for Lc 75, and large or bold text for Lc 60, with anything
under Lc 30 treated as invisible. That reframes contrast as a property of the typography as
well as of the palette, which is what practitioners had been doing informally for years.

The disagreements with WCAG 2 are where the argument lives, and they run both ways. Mid grey
text on black can pass WCAG 2 at AAA and score under Lc 50, because the old formula's
handling of dark backgrounds flatters them; white on a mid grey can fail 4.5:1 and score
comfortably above Lc 75. Neither result is a bug in the tool that produced it. They are two
models of the same phenomenon, one fitted to 1990s CRT measurements and one to more recent
readability research, and where they part company is usually worth looking at with your own
eyes.

Status matters here, so state it plainly: APCA is a candidate for WCAG 3, and WCAG 3 is a
working draft with no date. Nothing about APCA is normative, audits and procurement still
quote 4.5:1, and a page that satisfies APCA is not thereby conformant. The sane position is
to keep meeting the WCAG 2 thresholds you are actually held to and use Lc as the second
opinion that catches what a symmetric ratio cannot, particularly in dark themes. Some design
systems already publish both: Radix guarantees its text steps to Lc 60 and Lc 90 against a
step 2 background, which is a contrast promise written into the
[colour ramp](/color-ramp) rather than checked afterwards.