Vitamin D Unit Converter

The factor is right. The thing it is applied to is not one molecule.

A vitamin D result is a mass of 25-hydroxyvitamin D in a volume of blood. The United States reports it in nanograms per millilitre; the United Kingdom, Europe, Australia and Canada report the same blood in nanomoles per litre, which counts molecules instead. Getting from one to the other means dividing by the mass of one mole. One quirk worth knowing first: 1 ng/mL and 1 µg/L are the same quantity written two ways, so a report in µg/L needs no conversion at all.

25-hydroxyvitamin D3 — calcifediol, the form your skin makes and most supplements contain — is C27H44O2. PubChem (CID 5283731) reports its molecular weight to four significant figures as 400.6 g/mol. One ng/mL is 1 × 10⁻⁶ g/L; divide by 400.64 g/mol and you get 2.4960064 nmol/L. Round to four significant figures and you have the 2.496 that circulates everywhere.

We then did what this site's testosterone converter does: sourced the molar mass properly, from NIST's standard atomic weights, which are published as intervals rather than points because the isotopic composition of ordinary terrestrial material genuinely varies. Carbon is [12.0096, 12.0116], hydrogen [1.00784, 1.00811], oxygen [15.99903, 15.99977]. Run those through C27H44O2 and the molar mass is anywhere from 400.602 to 400.670 g/mol, so the factor is an interval too: 2.4958221 to 2.4962418, a span of 0.0168%.

2.496 lands inside that interval. The testosterone exercise found the universally printed 0.0347 sitting 0.075% ABOVE the top of its equivalent interval — outside it, and not a rounding of anything. Here there is no such finding, and manufacturing one would be dishonest. If you have been using 2.496, keep using it. One caveat on provenance, stated because the rest of this page is heavily cited and this figure is not: we could not open a primary source that actually prints 2.496. Not the Endocrine Society guideline, not its erratum, not the Institute of Medicine report, not CDC. Every one of those that states a conversion at all uses a flat 2.5. So 2.496 is derived here from the molar mass rather than cited, and it carries no reference in the list at the foot of this page.

Which brings up the second figure, because vitamin D is unusual in having two in wide circulation. The Endocrine Society's 2011 guideline writes its thresholds as "below 20 ng/ml (50 nmol/liter)" and, in the 2024 guideline, "30 ng/mL (75 nmol/L)". Those are multiplications by exactly 2.5. So is the erratum the Society published in December 2011 to correct the conversion of its insufficiency band: "The correct nmol conversion for the definition of vitamin D insufficiency is 52.5–72.5 nmol/L". Twenty-one and twenty-nine ng/mL times 2.496 are 52.416 and 72.384. A flat 2.5 is 0.16% above the derived factor and 0.151% above the top of the interval NIST permits — twice as far out as testosterone's 0.0347. At 30 ng/mL the entire disagreement between every factor in circulation is 0.125 nmol/L, which is 0.050 ng/mL. It does not matter. We say so rather than leaving you to wonder.

What does matter is that none of the above is quite what your laboratory measured. It did not measure 25-OH-D3. CDC states the definition plainly: "In patient care and public health assessments, vitamin D is defined as the sum of 25-hydroxyergocalciferol (25-hydroxyvitamin D2) and calcifediol (25-hydroxyvitamin D3)." Ergocalciferol's 25-hydroxy metabolite is C28H44O2 — one carbon heavier, 412.65 g/mol against 400.64 — so its conversion factor is 2.4233612, not 2.4960064. The two intervals do not overlap: D2's runs 2.42318 to 2.42359 and D3's 2.49582 to 2.49624. A mass concentration that is the sum of two different molecules cannot be converted to a molar concentration without knowing the split, and a standard US report does not print the split.

So every converter online, including this one, applies the D3 factor to a number that is a sum — and the error is 2.91% of whatever fraction of your total is D2. At 30 ng/mL the gap between the all-D3 and the all-D2 reading is 2.18 nmol/L, about 17 times the entire conversion-factor argument above. This is not a hypothetical population: the Endocrine Society's own guideline notes that "pharmaceutical preparations in the United States contain only vitamin D2", so anyone on prescription vitamin D is carrying a D2 fraction. The converter above lets you see what that assumption is worth. It will not ask you to guess your own split, and neither should anything else.

And all of that is still smaller than the assay. To be certified by the CDC Vitamin D Standardization-Certification Programme an assay must meet "the performance criterion of ±5.0% mean bias when compared to the CDC reference measurement procedure and an overall imprecision of <10%". At 30 ng/mL that permits a calibration anywhere from 28.50 to 31.50 ng/mL — 60 times the rounding argument and 3.4 times the vitamer span, from an assay that has passed. CDC is explicit that this is a calibration criterion and not a per-sample one: "Due to differences in test selectivity, measurements on individual samples can exceed the calibration bias." Its published list shows how often. Among the assays certified in Q4 2025, the 2 LC-MS/MS methods came within ±5% on 48% and 50% of individual samples; the 4 immunoassays managed 23%, 28%, 33% and 35%.

WHAT THIS PAGE WILL NOT TELL YOU. Whether your number is good. Two bodies published cut points and they did not agree. The Institute of Medicine's report brief calls 20 ng/mL "the level that is needed for good bone health for practically all individuals". The Endocrine Society's 2011 guideline suggested "vitamin D deficiency be defined as a 25(OH)D below 20 ng/ml, insufficiency as a 25(OH)D of 21–29 ng/ml, and sufficiency as a 25(OH)D of 30–100 ng/ml". They agree that below 20 is deficient and disagree about everything above it, and this page does not adjudicate between them — the same way /charts/lab-reference-range-chart prints A1C and LDL thresholds as decisions a committee took rather than facts about biology. One development is worth reporting because it is a fact rather than a judgement: in 2024 the Endocrine Society withdrew its own higher figure, writing that it "no longer endorses the target 25(OH)D level of 30 ng/mL (75 nmol/L) suggested in the previous guideline" and "no longer endorses specific 25(OH)D levels to define vitamin D sufficiency, insufficiency, and deficiency". The 30 ng/mL line is still printed on laboratory reports and on most of the internet. The IOM had already said why that happens: "the measurements, or cut-points, of sufficiency and deficiency used by laboratories to report results have not been set based on rigorous scientific studies, and no central authority has determined which cut-points to use. A single individual might be deemed deficient or sufficient, depending on the laboratory where the blood is tested."

Vitamin D conversion chart — ng/mL to nmol/L

Find the ng/mL figure from your US report on the left. The first column is the molar-mass answer; the second is what a flat ×2.5 gives; the last is what the same total would be if all of it were 25-OH-D2 rather than D3.

NO ROW HERE IS MARKED SUFFICIENT, INSUFFICIENT OR DEFICIENT, and that is deliberate: the two bodies that published cut points disagreed about everything above 20 ng/mL, and in 2024 the one that published the higher figure withdrew it. Compare the third and fourth columns. The third is the entire cost of using a flat 2.5 instead of the molar mass — never as much as 0.60 nmol/L anywhere on this table. The fourth is what the same number becomes if the total your laboratory reported was ergocalciferol-derived rather than cholecalciferol-derived, which is a separate question from rounding and a much larger one. The µg/L note under each row is not a conversion — a microgram per litre and a nanogram per millilitre are the same quantity. Every figure computed by the same function the converter above calls.

Vitamin D conversion chart — nmol/L to ng/mL

The other direction, for a report issued in SI units. Multiply nmol/L by the molar mass over a thousand, or by the printed 0.4 — the two do not quite agree, and the third column is by how much.

The 52.5 and 72.5 nmol/L rows are here because the Endocrine Society printed exactly those two numbers in its 2011 erratum — "The correct nmol conversion for the definition of vitamin D insufficiency is 52.5–72.5 nmol/L" — and a US reader needs them back in ng/mL to use them. Convert them with the molar mass and they come back as 21.03 and 29.05 ng/mL rather than the 21 and 29 they started from, which is what a 2.5 conversion does to a round trip through a different factor. The 50 and 75 nmol/L rows are the same story for the guideline's own parentheticals.

Every route to the conversion factor, compared

The same conversion derived seven ways. Four are molar-mass routes and agree to four decimal places; one is the four-figure factor in general use and sits among them; one is a flat 2.5 and does not; one is the wrong quantity entirely.

The first two rows are the bounds NIST's standard atomic weight intervals put on the molar mass of C27H44O2; everything between them is permitted by natural isotopic variation. THE HEADLINE IS A NEGATIVE RESULT: 2.496 is inside the interval, so it is right and you should keep using it. For comparison, this site's testosterone converter found the printed 0.0347 sitting 0.075% above the top of the equivalent testosterone interval — outside it. What IS outside here is the flat 2.5, by 0.151%, twice as far out as testosterone's; it is the figure the Endocrine Society guideline and erratum both use, and at 30 ng/mL it costs 0.050 ng/mL, which is nothing. The last row is the trap: PubChem's ExactMass, 400.334130642, is the monoisotopic mass — the mass of a single molecule in which every carbon is ¹²C — not a molar mass, and it gives 2.4979. One honest note on provenance: no primary source we could open actually prints 2.496, so the "Printed factor" row is derived here rather than cited, and it carries no reference below.

What the D2 fraction does to the conversion

Your report gives one total. CDC defines that total as 25-OH-D2 plus 25-OH-D3, and the two molecules do not share a factor. Each row is a different assumption about the split; the 0% row is what every converter online prints.

THIS IS NOT A TABLE TO LOOK YOURSELF UP IN. A standard US report prints one total and no fractions; only an assay that separates 25-OH-D2 from 25-OH-D3 can tell you your split, and the ordering clinician has to ask for it. The table is here to price the assumption rather than hide it. It is not an idle assumption: the Endocrine Society's guideline states that "pharmaceutical preparations in the United States contain only vitamin D2", so every patient on prescription vitamin D carries a D2 fraction of some size. 25-OH-D3 is C27H44O2 at 400.64 g/mol against 25-OH-D2's C28H44O2 at 412.65; their NIST-permitted factor intervals, 2.49582–2.49624 and 2.42318–2.42359, do not overlap. The full span at 30 ng/mL is 2.18 nmol/L, roughly 17 times the entire argument about 2.496 versus 2.5. PubChem CID 5710148 reports 25-OH-D2's molecular weight as 412.6.

What a CDC-certified vitamin D assay is allowed to report

The calibration bias the CDC Vitamin D Standardization-Certification Program permits an assay to carry and still be certified: ±5.0% mean bias against the CDC reference measurement procedure, with imprecision below 10%. Read the row for a true concentration; the range is what a passing assay may return.

Source: CDC VDSCP, Certified Total 25-hydroxyvitamin D Assays, which states that the listed assays "have successfully met the performance criterion of ±5.0% mean bias when compared to the CDC reference measurement procedure and an overall imprecision of <10%", and that "Only data for total 25-hydroxyvitamin D was used for certification". The last column is the point of the page: the tolerance a certified assay carries is roughly sixty times the entire conversion-factor argument and several times the vitamer span. THIS IS NOT AN ERROR BAR ON YOUR OWN RESULT and it is not a reference range — it is what the standardisation programme permits an assay's calibration. CDC is explicit that it is not a per-sample guarantee: "Due to differences in test selectivity, measurements on individual samples can exceed the calibration bias." An assay that is not enrolled is not bound by even this.

How often a certified assay actually hits ±5% on one sample

CDC publishes, for every certified assay, the share of the 40 certification samples that came within ±5% of the reference value. Every assay below passed certification. This is the column that shows what "certified" does and does not buy you.

CDC's own definition of the column: "'Individual samples pass rate' is the percentage of individual samples out of the 40 provided that met certification criteria of ±5% bias", and it says the figure "provides some information about the selectivity of a test that meets the calibration criteria". Selectivity is precisely the D2/D3 question this page is about. The 2 LC-MS/MS methods in the quarter managed 48% and 50%; the 4 immunoassays managed 23%, 28%, 33% and 35%. These are certification samples rather than patient samples, and a miss is a bias above 5% rather than a wrong result — but a reader arguing about the fourth decimal place of a conversion factor should see this table first. Transcribed from the Q4 2025 rows of the CDC list; assays certified in earlier quarters only are not shown.

The published cut points, in both units, with what happened to them

Two bodies drew lines and did not agree. Both are printed here with their source. This page does not say which is right, and no figure here is a target.

THIS IS NOT A REFERENCE RANGE AND MUST NOT BE READ AS ONE. Each row is one number from one named body, converted. The two agree that below 20 ng/mL is deficiency and disagree about everything above it; this page prints both and picks neither, the same way /charts/lab-reference-range-chart prints A1C and LDL thresholds as decisions rather than measurements. One row carries a status because a fact changed, not because we are taking a side: in 2024 the Endocrine Society wrote that it "no longer endorses the target 25(OH)D level of 30 ng/mL (75 nmol/L) suggested in the previous guideline" and "no longer endorses specific 25(OH)D levels to define vitamin D sufficiency, insufficiency, and deficiency", and in the same guideline suggested "against routine screening for 25(OH)D levels" in healthy adults. The 30 ng/mL line is nonetheless still printed on laboratory reports, which the IOM had anticipated: "no central authority has determined which cut-points to use. A single individual might be deemed deficient or sufficient, depending on the laboratory where the blood is tested."

Frequently asked questions

How do I convert vitamin D from ng/mL to nmol/L?

Multiply ng/mL by 2.496. A 25-hydroxyvitamin D of 20 ng/mL is 49.9 nmol/L; 30 ng/mL is 74.9 nmol/L; 50 ng/mL is 124.8 nmol/L. The factor is the molar mass of 25-hydroxyvitamin D3, 400.64 g/mol, rearranged as 1000 ÷ M. To go the other way multiply nmol/L by 0.40064, which most sources round to 0.4. A report in µg/L needs no conversion at all: 1 µg/L = 1 ng/mL exactly.

Is the 2.496 vitamin D conversion factor accurate?

Yes, and we checked it the hard way, expecting to find otherwise. Derived from the molar mass the factor is 2.4960064, and NIST's standard atomic weight intervals put it between 2.4958221 and 2.4962418 — so 2.496 is inside the permitted range, 0.00026% from the derived value. The same exercise on testosterone found the printed 0.0347 sitting 0.075% outside its interval. What is outside here is the flat 2.5 many sources use instead: 0.16% high, and at 30 ng/mL worth 0.050 ng/mL — still nothing. One honesty note: we could not open a primary source that actually prints 2.496, so this page derives it rather than citing it.

Why do two vitamin D converters give me different nmol/L?

Because two factors are in circulation. One is 2.496, the four-significant-figure rounding of the molar mass; the other is a flat 2.5, which is what the Endocrine Society's 2011 guideline uses when it writes "below 20 ng/ml (50 nmol/liter)" and what its published erratum uses when it corrects the insufficiency band to "52.5–72.5 nmol/L". At 30 ng/mL they give 74.88 and 75.00 nmol/L — a gap of 0.12 nmol/L. Neither difference is clinically meaningful, and if a converter's answer differs from yours by more than that, it is using a different factor again or you have entered a different unit.

Does the same conversion factor apply to vitamin D2 and vitamin D3?

No, and this is the largest error on any vitamin D converter including this one. CDC defines total 25-hydroxyvitamin D as "the sum of 25-hydroxyergocalciferol (25-hydroxyvitamin D2) and calcifediol (25-hydroxyvitamin D3)". 25-OH-D3 is C27H44O2 at 400.64 g/mol; 25-OH-D2 is C28H44O2 at 412.65 g/mol. Their factors are 2.4960 and 2.4234 — 2.91% apart, with non-overlapping ranges. Converting a total of 30 ng/mL gives 74.88 nmol/L if all of it is D3 and 72.70 nmol/L if all of it is D2, a span of 2.18 nmol/L — around 17 times the entire 2.496-versus-2.5 argument. Your report does not state the split, so every converter assumes all D3.

What is 30 ng/mL of vitamin D in nmol/L?

74.88 nmol/L from the molar mass, or 75.0 nmol/L if you use a flat 2.5, which is the figure most guidelines print. In µg/L it is simply 30, because a microgram per litre and a nanogram per millilitre are the same quantity. What 30 ng/mL means is a separate question this page does not answer: the Institute of Medicine put the level needed for bone health in practically everyone at 20 ng/mL, the Endocrine Society's 2011 guideline put sufficiency at 30–100 ng/mL, and in 2024 the Endocrine Society withdrew that figure.

What is a normal or optimal vitamin D level?

This page will not give you a number, because the two bodies that published one disagreed and one of them has since withdrawn its figure. The Institute of Medicine's report brief calls 20 ng/mL (49.9 nmol/L) "the level that is needed for good bone health for practically all individuals". The Endocrine Society's 2011 guideline suggested "deficiency be defined as a 25(OH)D below 20 ng/ml, insufficiency as a 25(OH)D of 21–29 ng/ml, and sufficiency as a 25(OH)D of 30–100 ng/ml". In 2024 the Endocrine Society stated it "no longer endorses the target 25(OH)D level of 30 ng/mL (75 nmol/L) suggested in the previous guideline" and "no longer endorses specific 25(OH)D levels to define vitamin D sufficiency, insufficiency, and deficiency", and suggested against routine screening in healthy adults. The IOM had already warned that laboratory cut-points "have not been set based on rigorous scientific studies" and that "a single individual might be deemed deficient or sufficient, depending on the laboratory where the blood is tested". Read your figure against the range printed by the laboratory that ran it and discuss it with the clinician who ordered it.

Why is my vitamin D different at two different labs?

Almost certainly the assay, not the arithmetic. To be certified, an assay must meet "±5.0% mean bias when compared to the CDC reference measurement procedure and an overall imprecision of <10%" — which at a true 30 ng/mL permits a calibration anywhere from 28.50 to 31.50 ng/mL, roughly 60 times the whole conversion-factor argument. And ±5% is a calibration criterion, not a per-sample one: CDC states that "measurements on individual samples can exceed the calibration bias". Its published list shows the share of the 40 certification samples each assay landed within ±5% on. In Q4 2025 the LC-MS/MS methods managed 48% and 50%; the immunoassays managed 23%, 28%, 33% and 35%. Every one of them passed certification.

Is vitamin D in µg/L the same as ng/mL?

Yes, exactly the same, with no factor involved — 1 µg/L = 1 ng/mL. A microgram is a thousand nanograms and a litre is a thousand millilitres, so the two cancel. A report reading 32 µg/L is 32 ng/mL, which is 79.9 nmol/L. This trips people up because every other unit pair on a lab report needs a factor, so readers assume this one must too and go looking for a number that is not there. The unit that does need converting is nmol/L, and it is 2.5 times larger in numerical value, which is why an unlabelled "30" is genuinely ambiguous.

Sources

1. [1]PubChem Compound Summary CID 5283731 (25-hydroxyvitamin D3) — MolecularFormula "C27H44O2", MolecularWeight "400.6", ExactMass "400.334130642"
2. [2]PubChem Compound Summary CID 5710148 (25-hydroxyvitamin D2) — MolecularFormula "C28H44O2", MolecularWeight "412.6", ExactMass "412.334130642"
3. [3]NIST Atomic Weights and Isotopic Compositions — standard atomic weights of carbon [12.0096,12.0116], hydrogen [1.00784,1.00811] and oxygen [15.99903,15.99977]
4. [4]CDC Vitamin D Standardization-Certification Program — "In patient care and public health assessments, vitamin D is defined as the sum of 25-hydroxyergocalciferol (25-hydroxyvitamin D2) and calcifediol (25-hydroxyvitamin D3)"
5. [5]CDC VDSCP, Certified Total 25-hydroxyvitamin D Assays — the ±5.0% mean bias and <10% imprecision criteria, the "Only data for total 25-hydroxyvitamin D was used for certification" note, and the individual-sample pass rates quoted on this page
6. [6]Holick MF, Binkley NC, Bischoff-Ferrari HA, et al. Evaluation, Treatment, and Prevention of Vitamin D Deficiency: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2011;96(7):1911–1930 — "deficiency be defined as a 25(OH)D below 20 ng/ml, insufficiency as a 25(OH)D of 21–29 ng/ml, and sufficiency as a 25(OH)D of 30–100 ng/ml"
7. [7]Erratum. J Clin Endocrinol Metab. 2011;96(12):3908, correcting Recommendation 1.2 of the 2011 vitamin D guideline — "The correct nmol conversion for the definition of vitamin D insufficiency is 52.5–72.5 nmol/L"
8. [8]Demay MB, Pittas AG, Bikle DD, et al. Vitamin D for the Prevention of Disease: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2024;109(8):1907–1947 — "the Endocrine Society no longer endorses the target 25(OH)D level of 30 ng/mL (75 nmol/L) suggested in the previous guideline"
9. [9]Institute of Medicine. Dietary Reference Intakes for Calcium and Vitamin D, Report Brief, November 2010 (revised March 2011) — 20 ng/mL is "the level that is needed for good bone health for practically all individuals", and laboratory cut-points "have not been set based on rigorous scientific studies"