What measured irradiance actually means — and why LED count doesn't
Last reviewed July 2026.
Every LED mask is sold on a count. 132 medical-grade LEDs. 236 LEDs. 648 LED lights for full-face coverage. The count is real, printed on the box, and easy to compare — and it tells you almost nothing about how much light reaches your skin. A count is a number of parts. The thing red-light therapy actually runs on is a dose.
The three numbers that matter
| Irradiance | Power density at the skin, in mW/cm² — how hard the light is hitting each square centimetre while the mask runs. This is what a spectrometer at the mask surface reads, and it's what ranks our table. |
| Fluence (dose) | Energy delivered over a session, in J/cm² — irradiance × time. The clinical literature's working range for skin work sits roughly in the single digits of J/cm² per session; the cited reviews treat ~5–9 as the sweet spot. One joule per cm² is one milliwatt per cm² held for 1,000 seconds. |
| Wavelength | Where the light sits on the spectrum. The studied bands for skin are red around 630–660 nm and near-infrared around 810–850 nm. Light outside the studied bands doesn't earn a rank no matter how strong it is. |
Why the count misleads
Two masks in our ranking make the point without any editorializing. The Therabody TheraFace advertises 648 LEDs — by far the biggest count in the table — and the cited review metered its red + near-infrared portion at 7 J/cm² over six minutes, which puts it third by delivered dose. The Omnilux Contour, with 132 diodes, metered 16 J/cm² over its ten-minute session — first. More, smaller, dimmer LEDs spread over the same face can deliver less light than fewer, stronger ones. The count is a bill of materials; irradiance is what arrives.
The sharper case is the claimed output. Qure advertises “70 mW/cm²” for the Q-Rejuvalight Pro; the cited review's spectrometer read 5–11 mW/cm², averaging around 8 — partly, the reviewer notes, because a thick silicone layer over the LEDs absorbs output before it ever reaches skin. A published spec is a claim. A meter reading is a measurement. Our ranking uses the second and shows the first muted beside it, so the gap stays visible.
How to read our numbers
Every figure in the ranking comes from Light Therapy Insiders’ spectrometer tests (Alex Fergus) — one tester, one instrument, measured at the mask surface, with the review behind each number linked on its row. We run no tests ourselves. Where a review prints the session dose rather than irradiance, we show that dose divided over its measured window — arithmetic, disclosed on the row. And where a mask has no measurement under that protocol, it waits in the unranked section: numbers metered by different testers at different distances don't share a ruler, and mixing them would quietly break the comparison the table exists to make. Delivered dose still varies with fit and wearing distance — a mask that hovers off your cheekbones delivers less there than any bench number suggests.
What this page is not
Not medical advice, and not an efficacy claim for any product. Whether a given dose does anything for a given face is a clinical question — red-light therapy is not a treatment for any disease, and a strong meter reading is not a promise. What a measured number does tell you is simpler and worth the price of the mask: whether the light the marketing sells is the light the device delivers. For whose FDA paper each “cleared” claim rides, see the clearance record.