๐Ÿ‡บ๐Ÿ‡ธ USC
Health & Fitness
๐Ÿ’‰

A1C Calculator & A1C Chart

Convert A1C to estimated average glucose (eAG) and back, with the A1C chart and blood sugar ranges

An A1C calculator turns a hemoglobin A1C (HbA1c) result into estimated average glucose (eAG), the mg/dL number your glucose meter uses. The ADAG study formula, eAG = 28.7 × A1C − 46.7, puts an A1C of 7% at 154 mg/dL (8.6 mmol/L). By the ranges CDC and NIDDK use, an A1C below 5.7% is normal, 5.7% to 6.4% is prediabetes and 6.5% or above is in the diabetes range.

โœ…

Last updated: November 10, 2026

Method: ADAG study regression (Nathan et al., Diabetes Care 2008), eAG (mg/dL) = 28.7 × A1C − 46.7, as in the eAG table of the ADA Standards of Care 2026; mmol/L = mg/dL ÷ 18, the study's own factor. The reverse direction solves the same line for A1C. IFCC units from the NGSP master equation. Ranges from CDC, NIDDK and the ADA: normal below 5.7%, prediabetes 5.7% to 6.4%, diabetes 6.5% or above.

Included: Both directions, mg/dL and mmol/L, percent and mmol/mol, the range of the result, every calculation step, an A1C chart from 4% to 14% and a blood sugar chart for fasting, 2-hour and random tests.

Not included: Diagnosis, the CGM-based glucose management indicator (GMI), children and pregnancy (both were excluded from the ADAG study), conditions that distort A1C and treatment advice. Screening information, not a diagnosis: talk to a clinician about your results.

๐Ÿ’‰ A1C and average glucose

What do you want to convert?
%

US labs report A1C in percent (NGSP). Many other countries report HbA1c in mmol/mol (IFCC).

๐Ÿ’‰ Estimated average glucose (eAG)

154mg/dL
Diabetes range ยท A1C 7.0%
eAG in mmol/L
8.6 mmol/L
A1C in IFCC units
53 mmol/mol

Diabetes range: 6.5% or above

6.5% or above is the diabetes range. One result is not a diagnosis: unless blood sugar is unmistakably high, the ADA requires two abnormal results, from two different tests or the same test repeated. If you already have diabetes, the ADA goal for many adults is an A1C below 7%; your clinician sets your own goal.

๐Ÿงฎ How this was calculated

eAG = 28.7 ร— 7.0 โˆ’ 46.7 = 154.2 mg/dL โ†’ 154 mg/dL

154.2 รท 18 = 8.57 mmol/L โ†’ 8.6 mmol/L

IFCC = (7.0 โˆ’ 2.152) รท 0.09148 = 53.0 โ†’ 53 mmol/mol

eAG line from the ADAG study (Nathan et al., 2008); 18 mg/dL of glucose equal 1 mmol/L, the factor behind the study's own mmol/L line. IFCC units from the NGSP master equation. Results are rounded half up.

๐Ÿ“Š Where this A1C falls

RangeA1Cmmol/moleAG mg/dL
NormalBelow 5.7%Below 39Below 117
Prediabetes range5.7% to 6.4%39 to 47117 to 139
Diabetes range6.5% or above48 or above140 or above

A1C ranges from CDC, NIDDK and the ADA Standards of Care 2026. The eAG column shows, rounded to whole mg/dL, the average glucose that matches each A1C range; it is not a cutoff for a fasting or random glucose test.

Screening information, not a diagnosis. Diabetes is diagnosed by a clinician from lab tests, usually two of them. Do not start, stop or change any medication based on this page; talk with your healthcare provider.

A1C calculator: the A1C chart, eAG and blood sugar ranges explained

The A1C test measures the percentage of your hemoglobin, the oxygen carrier in red blood cells, that has glucose attached. NIDDK notes that it is also called the hemoglobin A1C, HbA1c, glycated hemoglobin or glycohemoglobin test. Because red blood cells live about 120 days, the result reflects your average blood sugar over the past 2 to 3 months. US labs report it as a percentage; many other countries use mmol/mol.

Estimated average glucose (eAG) translates that percentage into mg/dL or mmol/L, the units of a home glucose meter. The American Diabetes Association and the American Association for Clinical Chemistry judged the correlation in the ADAG study (r = 0.92) strong enough to justify reporting eAG next to the A1C result when a clinician orders the test.

The A1C chart: A1C to estimated average glucose

Every row was computed with the same functions the calculator uses. The whole-number rows from 5% to 12% match the ADAG table that the ADA reprints as Table 6.1 of its Standards of Care; the half-point rows come from the same line.

A1C (%) mmol/mol eAG mg/dL eAG mmol/L Range
4.0% 20 68 3.8 Normal
4.5% 26 82 4.6 Normal
5.0% 31 97 5.4 Normal
5.5% 37 111 6.2 Normal
6.0% 42 126 7.0 Prediabetes range
6.5% 48 140 7.8 Diabetes range
7.0% 53 154 8.6 Diabetes range
7.5% 58 169 9.4 Diabetes range
8.0% 64 183 10.2 Diabetes range
8.5% 69 197 11.0 Diabetes range
9.0% 75 212 11.8 Diabetes range
9.5% 80 226 12.6 Diabetes range
10.0% 86 240 13.4 Diabetes range
10.5% 91 255 14.1 Diabetes range
11.0% 97 269 14.9 Diabetes range
11.5% 102 283 15.7 Diabetes range
12.0% 108 298 16.5 Diabetes range
12.5%* 113 312 17.3 Diabetes range
13.0%* 119 326 18.1 Diabetes range
13.5%* 124 341 18.9 Diabetes range
14.0%* 130 355 19.7 Diabetes range

* The ADAG table ends at 12%, so higher rows are extrapolated. Each full point of A1C adds 28.7 mg/dL to the estimated average, or 14.35 mg/dL per half point.

The formula behind the conversion

The calculator uses the regression line of the A1C-Derived Average Glucose (ADAG) study, published by Nathan and colleagues in Diabetes Care in 2008:

eAG (mg/dL) = 28.7 × A1C − 46.7 eAG (mmol/L) = eAG (mg/dL) ÷ 18 A1C (%) = (eAG in mg/dL + 46.7) ÷ 28.7 A1C (%) = 0.09148 × A1C (mmol/mol) + 2.152

The study followed 507 adults at 10 centers for 3 months: 268 with type 1 diabetes, 159 with type 2 diabetes and 80 without diabetes. Each collected about 2,700 glucose values from repeated continuous glucose monitoring and fingerstick profiles, and a straight line fit the data best (Rยฒ = 0.84). The paper's mmol/L version, 1.5944 × A1C − 2.594, is the mg/dL line divided by 18. The last line is the master equation the NGSP publishes to link US results to the IFCC reference method.

Older charts may show higher numbers. An earlier analysis of DCCT data used AG = 35.6 × A1C − 77.3, which puts 7% at 172 mg/dL instead of 154. NGSP notes that the ADAG line gives lower values with less scatter, and gives as the proposed explanation the far more frequent glucose measurements behind it.

Worked example: an A1C of 7%

Your lab report shows an A1C of 7.0%. Multiply by 28.7 to get 200.9, then subtract 46.7: the estimated average glucose is 154.2, shown as 154 mg/dL. Dividing 154.2 by 18 gives 8.57, or 8.6 mmol/L. In IFCC units, (7.0 − 2.152) ÷ 0.09148 = 53 mmol/mol. All three values match Table 6.1 of the ADA Standards of Care 2026, and 7.0% is the calculator's default input, so you can follow each step on screen.

The estimate has a spread. In the ADAG data, the 95% interval around 154 mg/dL ran from 123 to 185 mg/dL, so two people with the same A1C can have quite different averages.

Worked example: from a meter average to A1C

Your meter shows a 30-day average of 180 mg/dL. Switch the calculator to Glucose to A1C: (180 + 46.7) ÷ 28.7 = 7.90, so the estimated A1C is 7.90% (63 mmol/mol). Treat it as a rough guide, since a meter average depends on when you test. For data from a continuous glucose monitor (CGM), a different formula is standard; see the section on GMI below.

Normal, prediabetes and diabetes ranges

CDC, NIDDK and the ADA use the same three A1C ranges for adults who are not pregnant: below 5.7% is normal, 5.7% to 6.4% (39 to 47 mmol/mol) is prediabetes, and 6.5% (48 mmol/mol) or above is diabetes. The ADA stresses that risk is continuous: it extends below the prediabetes range and becomes disproportionately greater at its upper end. NIDDK makes the same point for A1C: within 5.7% to 6.4%, the higher the A1C, the greater the risk of diabetes.

The diabetes cutoff is a diagnostic threshold, not a verdict from one test. Unless blood sugar is unmistakably high, as with classic symptoms or a hyperglycemic crisis, the ADA requires two abnormal results: two different tests done at the same time, such as A1C and fasting glucose, or the same test at two different times. A person with two A1C results of 6.5% or above still has diabetes if fasting glucose is below 126 mg/dL. Results close to a threshold should lead to a repeat test in 3 to 6 months.

Prediabetes: an A1C of 5.7% to 6.4%

Say a routine checkup returns an A1C of 5.9%. The eAG is 28.7 × 5.9 − 46.7 = 122.6, about 123 mg/dL (6.8 mmol/L), and the IFCC value is 41 mmol/mol. Under the ADA criteria, the next test is due in a year. The risk can be lowered: in the Diabetes Prevention Program, a lifestyle program with a goal of losing 7% of body weight and 150 minutes of activity a week cut the chance of developing type 2 diabetes by 58% compared with placebo, and by 71% in people 60 and older; metformin cut it by 31%. After 15 years, the lifestyle group still developed diabetes at a 27% lower rate. CDC puts the weight loss that helps at around 5% to 7% of body weight; the Weight Loss Calculator shows what that means for you.

Blood sugar chart: fasting, 2-hour and random tests

Blood glucose tests have their own cutoffs. Here is the NIDDK chart, with the mmol/L values from the ADA Standards of Care:

Test Normal Prediabetes Diabetes
A1Cbelow 5.7%5.7% to 6.4%6.5% or above
Fasting plasma glucose (no calories for at least 8 hours)99 mg/dL or below (below 5.6 mmol/L)100 to 125 mg/dL (5.6 to 6.9 mmol/L)126 mg/dL or above (7.0 mmol/L or above)
2 hours into a 75 g oral glucose tolerance test139 mg/dL or below (below 7.8 mmol/L)140 to 199 mg/dL (7.8 to 11.0 mmol/L)200 mg/dL or above (11.1 mmol/L or above)
Random plasma glucoseno cutoffno cutoff200 mg/dL or above (11.1 mmol/L) with symptoms

These cutoffs apply to plasma glucose measured by a lab, not to fingerstick readings, and the ADA says there is not yet enough evidence to use CGM to screen for or diagnose diabetes. The glucose tolerance test also needs preparation: the ADA advises eating at least 150 g of carbohydrates a day on the 3 days before the test, because cutting carbohydrates beforehand can falsely raise the 2-hour value.

Blood sugar targets if you have diabetes

The ADA Standards of Care 2026 list these goals for many nonpregnant adults with diabetes:

Measure Goal
A1Cbelow 7% (53 mmol/mol)
Glucose before meals80 to 130 mg/dL (4.4 to 7.2 mmol/L)
Peak glucose after meals (1 to 2 hours after starting)below 180 mg/dL (10.0 mmol/L)
CGM time in range, 70 to 180 mg/dLmore than 70%
CGM time below 70 mg/dLless than 4%

Low blood sugar has its own scale. Level 1 hypoglycemia is a glucose below 70 mg/dL (3.9 mmol/L) but at or above 54 mg/dL (3.0 mmol/L), level 2 is below 54 mg/dL, and level 3 is a severe event with altered mental or physical status that needs another person's help, whatever the number. Goals are individual: the ADA allows lower A1C goals, such as below 6.5%, for people in good health with a low risk of hypoglycemia, and suggests below 8.0% for older adults with complex or intermediate health. In pregnancy, A1C runs slightly lower because red blood cells turn over faster, and the goals are tighter: fasting glucose below 95 mg/dL, and below 140 mg/dL one hour or 120 mg/dL two hours after meals.

When A1C can be misleading

A1C is an indirect measure, and the ADA lists situations in which it drifts away from true average glucose:

  • Red blood cell turnover: anemia, iron deficiency, blood loss, transfusion, hemolysis, glucose-6-phosphate dehydrogenase (G6PD) deficiency, erythropoietin treatment and splenectomy. NIDDK notes that very low iron can produce a falsely high A1C.
  • Other illness: kidney failure and dialysis, cirrhosis and HIV.
  • Pregnancy, which lowers A1C slightly.
  • Hemoglobin variants such as sickle cell trait, which interfere with some lab methods. In sickle cell disease (HbSS) and other homozygous variants, A1C cannot be measured at all.

In these cases the ADA recommends diagnosing diabetes with plasma glucose tests, and monitoring with glucose meters, CGM or alternative markers such as fructosamine. A persistent, large gap between your A1C and your glucose readings is a good reason to ask about these factors.

eAG, GMI and your meter average

Three numbers often get mixed up:

  • eAG is calculated from a lab A1C with the ADAG line. It describes the average glucose that typically goes with that A1C.
  • GMI, the glucose management indicator, goes the other way: CGM software estimates an A1C-like value from at least 10, preferably 14 or more, days of sensor data with GMI (%) = 3.31 + 0.02392 × mean glucose in mg/dL (Bergenstal et al., 2018).
  • A meter average is the mean of the fingersticks you happened to take, so it depends on when you test. NGSP notes that fasting values underestimate average glucose more and more as A1C rises.

GMI was fitted on newer CGM data, so it does not match the reversed ADAG line. At an average of 150 mg/dL the two are close (6.85% and 6.90%), but at 200 mg/dL the ADAG line gives 8.60% and GMI 8.09%, and at 100 mg/dL 5.11% against 5.70%. The ADA calls GMI an approximation of A1C, not always equal to it, and notes that it is unclear how well the 2008 ADAG estimates carry over to modern CGM systems.

How to use this calculator

  1. Choose the direction. A1C to glucose for a lab result, Glucose to A1C for an average from your meter or CGM.
  2. Pick the unit. US labs report A1C in percent; switch to mmol/mol for results from other countries. Glucose works in mg/dL or mmol/L. Switching shows your entry in the other unit, and the result keeps using what you typed.
  3. Type the value. A1C is accepted from 3% to 20% and glucose from 40 to 500 mg/dL. A number outside that span, or an A1C below 20 mmol/mol, usually means the wrong unit, and the calculator says so.
  4. Read the result and its range, then check the steps: the calculation is shown line by line, so you can repeat it by hand.

Who should get tested, and how often

The ADA recommends screening all adults from age 35, and testing earlier in adults with overweight or obesity (a BMI of 25 or more, or 23 or more with Asian ancestry) plus another risk factor, such as a parent or sibling with diabetes, high blood pressure, physical inactivity or a history of heart disease. Normal results are repeated at least every 3 years, prediabetes every year, and people with diabetes check at least twice a year. CDC estimates that 40.1 million people in the US had diabetes in 2023, 11.0 million of them undiagnosed, and that 115.2 million adults have prediabetes. To see where your weight stands, use the BMI Calculator.

Limitations

The ADAG study excluded children, pregnant women and people with conditions that disturb A1C, such as anemia, hemoglobin variants, kidney or liver disease and recent blood loss, so the conversion applies to adults without those conditions. 83% of the participants were non-Hispanic White, and 98.5% of the samples had an A1C between 4% and 10%. The reverse conversion does not turn a meter average into a lab result. Most importantly, this page gives screening information, not a diagnosis: diabetes is diagnosed by a clinician from lab tests.

Related calculators

Sources

โš ๏ธ Common mistakes & edge cases

Comparing eAG with the fasting cutoff

An A1C of 6.5% matches an eAG of about 140 mg/dL, yet the fasting cutoff for diabetes is 126 mg/dL. Both are right: one is an around-the-clock average, the other a single value after at least 8 hours without food. Do not judge a fasting reading against the eAG chart, or the other way round.

Expecting eAG to equal your meter average

NIDDK warns that eAG will not match daily readings. A meter average leans on when you test, and even in the ADAG study an A1C of 7% went with averages from 123 to 185 mg/dL.

Reading mmol/mol as a percentage

An HbA1c of 48 on a report from outside the US means 48 mmol/mol, which is 6.5%, not 48%. The calculator rejects A1C values above 20%, flags values below 20 mmol/mol, and in both cases suggests switching the unit.

Treating one result as a diagnosis

Without unmistakable symptoms, the ADA requires two abnormal results. A single A1C of 6.6% is a reason to repeat or add a test, not a diagnosis, and an A1C estimated from a meter average is not a test at all.

Ignoring what distorts A1C

Anemia, iron deficiency, recent blood loss or a transfusion, kidney failure, pregnancy and some hemoglobin variants can push A1C away from true average glucose. In those cases the ADA relies on plasma glucose tests for diagnosis.

Note: This calculator converts numbers; it does not diagnose diabetes or prediabetes. Talk to your healthcare provider about any A1C or glucose result, and seek care promptly for very high readings with classic symptoms such as frequent urination, extreme thirst or unexplained weight loss.

❓ Frequently asked questions

What is a normal A1C level?

Below 5.7%. CDC and NIDDK use three ranges: below 5.7% is normal, 5.7% to 6.4% is prediabetes and 6.5% or above is diabetes. In mmol/mol, the unit used in many other countries, normal is below 39. For people who already have diabetes the question is a different one: the ADA goal for many nonpregnant adults is an A1C below 7% (53 mmol/mol), adjusted by a clinician to the person.

What A1C level is prediabetes?

An A1C of 5.7% to 6.4% (39 to 47 mmol/mol) is the prediabetes range. It matches an estimated average glucose of about 117 to 139 mg/dL. NIDDK notes that within this range, the higher the A1C, the greater the risk of diabetes. Once you are in this range, the ADA recommends testing again every year.

What A1C level means diabetes?

An A1C of 6.5% (48 mmol/mol) or higher is in the diabetes range, which corresponds to an estimated average glucose of 140 mg/dL (7.8 mmol/L) or more. One result is not a diagnosis. Unless blood sugar is unmistakably high, the ADA requires two abnormal results: two different tests, such as A1C and fasting glucose, or the same test repeated at a later time.

How do I convert A1C to average blood sugar?

Use the ADAG formula: eAG (mg/dL) = 28.7 ร— A1C โˆ’ 46.7. For an A1C of 7%, 28.7 ร— 7 = 200.9, and 200.9 โˆ’ 46.7 = 154.2, so about 154 mg/dL. Divide by 18 for mmol/L: 8.6 mmol/L. The formula comes from a 2008 study of 507 adults who used continuous glucose monitors and fingerstick checks for 3 months, and the ADA prints the same values in its Standards of Care.

What is hemoglobin A1C (HbA1c)?

Hemoglobin A1C, also written HbA1c or A1C and called glycated hemoglobin, is the percentage of hemoglobin in your red blood cells that has glucose attached. Red blood cells live about 120 days, so the result reflects your average blood sugar over the past 2 to 3 months, weighted toward the most recent weeks. You do not need to fast for the test.

What is a normal blood sugar level?

For a fasting test, 99 mg/dL (5.5 mmol/L) or below is normal, 100 to 125 mg/dL is prediabetes and 126 mg/dL (7.0 mmol/L) or above is diabetes. Two hours into an oral glucose tolerance test, 139 mg/dL or below is normal, 140 to 199 mg/dL is prediabetes and 200 mg/dL or above is diabetes. A random test has no normal cutoff; 200 mg/dL (11.1 mmol/L) or above together with symptoms of high blood sugar points to diabetes.

What blood sugar should I aim for if I have diabetes?

The ADA goals for many nonpregnant adults are 80 to 130 mg/dL (4.4 to 7.2 mmol/L) before meals, below 180 mg/dL (10.0 mmol/L) at the peak after meals, measured 1 to 2 hours after you start eating, and an A1C below 7%. People using a CGM aim for more than 70% of the time between 70 and 180 mg/dL. Goals are individual and set with your clinician: for older adults with complex health, for example, the ADA suggests an A1C below 8.0%.

Why doesn't my eAG match my glucose meter average?

NIDDK points out that eAG will not match daily readings because it is a long-term average. Meter checks are snapshots, often taken fasting or before meals, and NGSP notes that fasting glucose underestimates the average more and more as A1C rises. The ADAG data also scatter: an A1C of 7% went with average glucose from 123 to 185 mg/dL (95% interval). The ADA calls for a closer look when meter or CGM values and A1C disagree consistently and substantially.

How do I convert HbA1c from mmol/mol to percent?

Use the NGSP master equation: percent = 0.09148 ร— mmol/mol + 2.152. For 48 mmol/mol, 0.09148 ร— 48 = 4.39, plus 2.152 gives 6.54, which a lab reports as 6.5%. The other way round, mmol/mol = (percent โˆ’ 2.152) รท 0.09148, so 7% is 53 mmol/mol and 5.7% is 39 mmol/mol. Switch the unit in the calculator to convert either way.

Can an A1C result be wrong?

Yes. Several conditions can push A1C away from your true average glucose, including anemia, iron deficiency, blood loss, transfusions, hemolysis, kidney failure, liver disease and pregnancy. Some hemoglobin variants, such as sickle cell trait, interfere with certain lab methods, and in sickle cell disease (HbSS) A1C cannot be measured at all. In these situations the ADA recommends diagnosing diabetes with plasma glucose tests instead.

How often should A1C be checked?

Without diabetes, the ADA recommends screening from age 35, or earlier for adults with overweight and another risk factor, and repeating it at least every 3 years when results are normal. With prediabetes, test every year. With diabetes, the ADA advises checking glycemic status at least twice a year, and about every 3 months if you are not meeting your goals or your treatment has changed.

Can I lower my risk if my A1C is in the prediabetes range?

Yes. In the Diabetes Prevention Program, a lifestyle program aimed at losing 7% of body weight and exercising 150 minutes a week lowered the chance of developing type 2 diabetes by 58% compared with placebo, and by 71% in people 60 and older. CDC's lifestyle change program cites the same figures and puts the weight loss that helps at around 5% to 7% of body weight.

๐Ÿ’ก Good to know

No fasting needed

You do not have to fast for an A1C test, so blood can be drawn at any time of day (NIDDK). Unlike a glucose test, A1C is also unaffected by a recent meal, stress, illness or exercise (ADA).

The last month counts most

A1C covers about 3 months, but the most recent 30 days weigh substantially more than the days 90 to 120 back (NGSP). A big change in blood sugar shows up well before a full red cell lifespan has passed.

There is no safe side of a cutoff

The ADA notes that risk is continuous: it extends below the prediabetes range and rises disproportionately at its upper end. 5.6% is not risk-free, and 6.4% deserves more attention than 5.7%.

Related Calculators