Stevia vs Erythritol
Stevia and erythritol are the most common pairing in the sugar-free aisle, and often they are the same product: a tub labeled “stevia” is mostly bulking agent by weight, and that bulking agent is commonly erythritol. So the useful question is less “which one?” than “which does what?”. Stevia brings intense sweetness from milligrams of purified leaf compounds. Erythritol brings bulk, a cooling sensation and most of the weight. They also differ sharply in how settled their science is: stevia is the best-characterized sweetener we cover, while erythritol carries a cardiovascular question that is real and still open.
At a Glance
Every row below comes from our two sourced records, the same ones the cards at the foot of this page show. Each figure links to the document it came from and the date we last checked it; open “Note” for the qualifications that go with it. Where a record has nothing sourced for a row, the cell says so rather than filling the gap.
| Attribute | Stevia | Erythritol |
|---|---|---|
| Sweetness | 200–400× sweeter than sugarUS Food and Drug Administration regulatory checked 30 Aug 2026NoteA MULTIPLE, not a fraction. FDA's figure for steviol glycosides as a class. Individual glycosides differ: stevioside is the least potent and rebaudioside A among the most, so a reformulation that swaps Reb A for Reb M changes potency without changing the word "stevia" on the label. | 60–80% of sugarEFSA Panel on Food Additives and Flavourings, EFSA Journal… regulatory checked 29 Aug 2026NoteSpecification band. Measured potency is concentration-dependent: 0.72 / 0.75 / 0.84 against 5 / 10 / 15% w/v sucrose (Wee 2018); 0.50-0.78 across 3-20% (Fujimaru 2012). |
| Calories | No sourced entry | 0 kcal per gramElectronic Code of Federal Regulations regulatory checked 29 Aug 2026 |
| Net carbs | Subtracts in full — none of its grams countUS Food and Drug Administration regulatory checked 30 Aug 2026NoteApplies to the PURIFIED GLYCOSIDES ONLY, used at milligram levels. As with monk fruit, a retail "stevia" product is mostly bulking agent by weight — commonly erythritol, dextrose or maltodextrin — and dextrose and maltodextrin are NOT low-glycaemic. Compute the carbohydrate from the carrier, and note that the carrier differs between brands. | Subtracts in full — none of its grams countElectronic Code of Federal Regulations regulatory checked 29 Aug 2026NoteFDA assigns erythritol 0 cal/g, uniquely among polyols (all others 2.0-3.0). Glycaemic index 0. |
| Blood glucose | Low-certainty evidence of reduced fasting glucose, strongest under 120 days of useZare M, Zeinalabedini M, Ebrahimpour-Koujan S, Bellissimo N, Azadbakht… peer reviewed checked 31 Aug 2026NoteZare et al. 2024, meta-analysis of 26 RCTs, 1,439 participants (PROSPERO CRD42023414411). Stevia supplementation significantly reduced fasting glucose, with the effect most consistent in trials shorter than 120 days and in participants with higher BMI, diabetes or hypertension. NO significant effect on insulin concentration or HbA1c — the fasting-glucose finding does not generalise to those two measures. Authors rate the overall certainty of evidence LOW and call for more RCTs at higher stevia doses. Read alongside health_claim_status below: a real, specific, but modest and short-duration signal — not evidence of durable metabolic benefit. | 0 (glucose = 100)Livesey G, Nutrition Research Reviews 16:163-191 peer reviewed checked 29 Aug 2026NoteMean 0, SD 17 across 4 studies (glucose = 100). Insulinaemic index 2. The SD is large relative to the mean: this is "no detectable effect", not a precise zero. |
| Acceptable daily intake | 4 mg per kg body weight per dayEFSA Panel on Food Additives and Flavourings, EFSA Journal… regulatory checked 30 Aug 2026NoteExpressed as STEVIOL EQUIVALENTS, not as grams of product — a conversion step every formulator has to do and many skip. Set by JECFA at its 69th meeting in 2008 (corrected 2026-08-31 from an earlier "2009" taken second-hand; the JECFA database itself says 2008, when the temporary designation was removed) on a NOEL of 970 mg/kg bw/day from a long-term stevioside study with a 100-fold safety factor, and independently by EFSA in 2010 from a 2-year rat carcinogenicity study. In November 2024 EFSA assessed an applicant request to raise it to 6 or 16 mg/kg bw/day and refused: "there is insufficient justification to increase the current ADI." | EU 0.5 g per kg body weight per dayEFSA Panel on Food Additives and Flavourings, EFSA Journal… regulatory checked 29 Aug 2026NoteEFSA pooled NOAELs for diarrhoea from 10 human studies: range 0.51-1.0 g/kg bw, median 0.75. EFSA adopted the lower bound, 0.5 g/kg bw/day, as its first numerical ADI for erythritol (Dec 2023) and found children's 95th-percentile intake already exceeds it. |
| Digestive tolerance | No sourced entry | 0.5–1 g/kg in one sitting · EU limit 0.5 g/kg per dayEFSA Panel on Food Additives and Flavourings, EFSA Journal… regulatory checked 29 Aug 2026NoteEFSA pooled NOAELs for diarrhoea from 10 human studies: range 0.51-1.0 g/kg bw, median 0.75. EFSA adopted the lower bound, 0.5 g/kg bw/day, as its first numerical ADI for erythritol (Dec 2023) and found children's 95th-percentile intake already exceeds it. |
| Browning | No sourced entry | NoneEuropean Commission regulatory checked 29 Aug 2026NotePolyols do not participate in Maillard browning; the additive specification caps reducing substances at 0.3% as D-glucose. |
| Bulk contribution | NoneUS Food and Drug Administration regulatory checked 30 Aug 2026NoteHigh-intensity; contributes no volume, structure, browning or mouthfeel. Anything cup-for-cup is mostly carrier. | BulkingEFSA Panel on Food Additives and Flavourings, EFSA Journal… regulatory checked 29 Aug 2026NoteUsed as a bulk sweetener, but a 1:1 weight swap under-sweetens because it delivers only 60-80% of sucrose sweetness. |
| Recrystallisation | No sourced entry | HighTyapkova et al. peer reviewed checked 29 Aug 2026NoteLow solubility relative to sucrose; recrystallises on cooling, which is the mechanism behind the gritty or sandy texture in baked goods and frozen desserts. |
| United States | GRAS by notification, high-purity onlyUS Food and Drug Administration regulatory checked 30 Aug 2026NoteHARD PURITY THRESHOLD. FDA has evaluated many GRAS notices for "high purity (95% minimum purity) steviol glycosides" including Reb A, stevioside and Reb D. But: "The use of stevia leaf and crude stevia extracts is not considered GRAS, and their import into the U.S. is not permitted for use as sweeteners" — Import Alert 45-06. A whole-leaf or crude-extract product is not a lower grade of the same legal thing; it is a different regulatory category and it is not permitted. | GRAS by notificationUS Food and Drug Administration regulatory checked 27 Oct 2026NoteSix FDA "no questions" letters, per FDA's GRAS Notice Inventory (checked 2026-10-27): GRN 76 (closed 2001-09-11), 208 (2007-01-25), 382 (2011-11-21), 401 (2012-03-22), 789 (Cargill, 2019-02-20) and 1296 (Shandong Starlight So True, 2026-04-07), the last for erythritol "produced through fermentation of glucose by Yarrowia lipolytica". CORRECTED 2026-10-27: this note previously said six letters ran from GRN 76 through GRN 789, which is five. The production organism now varies by supplier, so prose naming only Moniliella is incomplete. Intended uses in GRN 1296 exclude infant formula and USDA-regulated products. |
| European Union | Authorised, E 960a-dEFSA Panel on Food Additives and Flavourings, EFSA Journal… regulatory checked 30 Aug 2026NoteAuthorised across 32 food categories. CORRECTED 2026-08-31: this note previously omitted E 960b, one of four group members confirmed directly against the EU Food Additives register. E 960a is steviol glycosides obtained by water extraction from stevia leaf; E 960b is steviol glycosides from fermentation; E 960c covers enzymatically produced variants including rebaudioside M, D and AM; E 960d is glucosylated steviol glycosides. Route of manufacture therefore changes the E number while the front label still says stevia. | Authorised, E 968EFSA Panel on Food Additives and Flavourings, EFSA Journal… regulatory checked 29 Aug 2026NoteVERIFIED DIRECTLY against the EU Food Additives register, 2026-08-31 — do not confuse the two authorisation layers. The substance-specific listing for erythritol shows only ONE category (14.1.4, flavoured drinks, as a flavour enhancer, ML 16000 mg/l, energy-reduced/no-added-sugar only, under Reg (EU) 2015/1832) — reading only that would wrongly suggest erythritol is authorised for almost nothing. The real authorisation is broader: erythritol is a member of GROUP IV, POLYOLS, alongside sorbitol (E420), mannitol (E421), isomalt (E953), maltitol (E965) and xylitol (E967), and Group IV carries its own quantum satis authorisation across roughly 30 food categories (energy-reduced or no-added-sugar foods) under Reg (EU) No 1129/2011. EFSA re-evaluation Dec 2023 established an ADI of 0.5 g/kg bw/day and declined to remove the laxative warning required for foods containing more than 10% erythritol. |
What Each One Does
FDA puts steviol glycosides, the sweet compounds purified from the stevia leaf, at 200 to 400 times sweeter than sugar. The individual glycosides differ in potency, so a product that switches from one to another changes how sweet it is without changing the word “stevia” on the label. Used at milligram levels, they add no volume, structure, browning or mouthfeel.
Erythritol is a sugar alcohol used as a bulk sweetener. Its specification band is 60–80% of sugar’s sweetness, and measured sweetness rises with concentration, so a one-for-one swap by weight under-sweetens. It has the strongest cooling effect of the common sugar alcohols, because it absorbs heat as it dissolves. It does not brown, since sugar alcohols do not take part in the Maillard reaction, and its low solubility means it recrystallizes as foods cool, which is the gritty texture people notice in erythritol-sweetened baked goods and frozen desserts.
Put together, the two cover each other’s gaps: erythritol supplies the bulk and stevia supplies the sweetness erythritol lacks. That is why the combination is everywhere.
Calories, Carbs and Blood Sugar
FDA assigns erythritol 0 kcal/g, the only sugar alcohol with a zero factor, and its glycemic index is 0 (averaged across four studies, with enough scatter that “no detectable effect” is the honest reading). Its grams count in Total Carbohydrate on a US label but not in Total or Added Sugars.
Purified steviol glycosides are not broken down or absorbed in the upper gut; bacteria in the colon cleave them to steviol. At milligram use levels they subtract in full from net carbs. The carrier is another matter. Retail stevia products are bulked with erythritol, dextrose or maltodextrin depending on the brand, and dextrose and maltodextrin are ordinary, rapidly digested carbohydrate. Read the ingredient list.
Stevia has the stronger direct evidence on glucose. A 2024 meta-analysis of 26 trials with 1,439 participants found stevia lowered fasting glucose, most consistently in trials shorter than 120 days and in people with diabetes, higher BMI or hypertension. It found no effect on insulin or HbA1c, and its authors rate the certainty low. Stevia is well evidenced as a way to replace sugar; evidence that it improves metabolic health on its own is not.
Intake Limits
The two limits are not in the same units, and neither is in grams of the product you buy.
Stevia’s acceptable daily intake is 4 mg per kg of body weight a day, expressed as steviol equivalents, not as grams of product. JECFA set it in 2008 and EFSA reached the same figure independently in 2010. In 2024 EFSA refused a request to raise it, and found that a proposed extension into baked goods would push toddlers’ high-end intake (6.9 mg/kg a day at the 95th percentile) further above it.
Erythritol’s limit comes from the gut. EFSA pooled ten human studies with diarrhea thresholds between 0.5 and 1 g per kg of body weight in a single dose, and in 2023 adopted the low end, 0.5 g/kg a day, as erythritol’s first numerical ADI. It found that children’s high intakes already exceed it, and it kept the EU’s laxative warning for foods with more than 10% erythritol. For a 70 kg adult, a single 30 g serving is about 86% of that daily limit.
The Cardiovascular Question
This is where the two part company. Stevia has no comparable open signal in our record. Erythritol does, and it has to be stated plainly. A 2023 study in Nature Medicine linked higher blood erythritol to a higher risk of major adverse cardiac events. The ARIC study of 4,006 older adults free of heart disease linked it to heart-failure hospitalization (HR 1.18), cardiovascular death (HR 1.21) and death from any cause (HR 1.19). Three genetic (Mendelian randomization) studies published in 2025 and 2026 found higher genetically predicted erythritol going with higher odds of stroke or coronary disease.
The caveats are just as real. All of this measures erythritol in the blood, not in the diet, and the body makes erythritol from glucose. In the Nurses’ Health Study the association faded once diabetes was accounted for, while the same-size signal for mannitol and sorbitol, which almost nobody eats in quantity, did not. FDA’s review of the 2023 study noted it did not adjust for kidney function, which also raises blood erythritol. The authors of the 2026 genetic study say their result should not be read as evidence that eating erythritol causes stroke. No regulator has restricted erythritol. It is an unresolved signal that is still moving, not a settled verdict either way.
Regulatory Status
Stevia’s position is the most settled of the two. In the US, high-purity (at least 95%) steviol glycosides are GRAS by notification, but stevia leaf and crude leaf extracts are not GRAS and may not be imported for use as sweeteners (Import Alert 45-06). In the EU, steviol glycosides are authorized as E 960a–d across 32 food categories; the four numbers distinguish leaf extraction, fermentation, enzymatic production and glucosylation.
Erythritol is GRAS by notification in the US, with six FDA “no questions” letters between 2001 and 2026, and is authorized in the EU as E 968, under the group authorization for polyols.
Which Should You Choose
For drinks, stevia on its own does the job with no carbohydrate and no digestive load; choose a liquid or a product whose carrier you have checked. For baking, you need bulk, and erythritol or an erythritol–stevia blend provides it, with the trade-offs above: no browning, a cooling note and a gritty texture once cooled. If you eat erythritol in quantity, keep the 0.5 g/kg daily limit in mind, and if you already have heart disease or strong risk factors, the open research question is reasonable to discuss with a doctor.
The Bottom Line
Stevia is the better-characterized of the two: an ADI two bodies agree on, full EU authorization and a settled US position, though no proven benefit beyond replacing sugar. Erythritol is the one that makes a stevia product usable by the cup, with a zero-calorie, zero-GI profile, a digestive limit lower than most people assume, and a cardiovascular signal that is associated, unexplained and not shown to be caused by diet. When you buy a “stevia” tub, check how much of it is erythritol, because that is most of what you are eating.
New comparisons and corrections, as they land.
This page is educational information, not medical or dietary advice. Nutrition and regulatory details change over time — verify against the cited primary sources before relying on them for formulation or health decisions.