Allulose vs Monk Fruit - Zero Sugar AI

Allulose and monk fruit turn up on the same shelf and often in the same recipe, but they are not two versions of one thing. Allulose is a rare sugar used by the spoonful: it has bulk, it browns, and it is about 70% as sweet as table sugar. Monk fruit extract is a high-intensity sweetener used by the milligram: 100 to 250 times sweeter than sugar, with no bulk at all, which is why a tub labeled “monk fruit” is mostly something else by weight. They have one surprising thing in common. Neither is authorized in the European Union, for two quite different reasons.

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	Allulose	Monk Fruit
Sweetness	70% of sugarXie X, Yu L, Lin Q, Huang D; Journal… peer reviewed checked 30 Aug 2026 Note Commonly cited as approximately 70% of sucrose. Convergent across Xie 2024, Han 2018 and the comments FDA received on the 2014 proposed rule — but this is a repeated figure, not a specification band or a sensory measurement, and no concentration dependence has been characterised the way it has for erythritol.	100–250× sweeter than sugarUS Food and Drug Administration regulatory checked 30 Aug 2026 Note A MULTIPLE, not a fraction — see the header warning. CORRECTED 2026-08-30 from an earlier 100-400 band: FDA gives 100-250 for the EXTRACT ("SGFE, depending on the mogroside content, is reported to be 100 to 250 times sweeter than sugar"), while the ~250 figure in the literature describes PURE mogroside V. Those are not in conflict once you separate extract from active compound, and the extract figure is the one that matters commercially. Potency tracks mogroside V content, which varies 2%-95% between grades.
Calories	0.4 kcal per gramUS FDA, Center for Food Safety and Applied Nutrition regulatory checked 30 Aug 2026 Note ENFORCEMENT DISCRETION, not a rule. FDA states that under 21 CFR 101.9(c)(1)(B) "a caloric value of 4 kcal/g must currently be used for allulose because it is a carbohydrate", and that it merely intends not to enforce that pending rulemaking. Two citizen petitions asked for 0.4 and 0.2; FDA chose the higher figure because it "would not underestimate the caloric contribution".	Negligible at use levelsUS FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026 Note Deliberately not stated as "0 kcal/g". No authority has assigned monk fruit extract a caloric factor the way FDA assigned erythritol 0 and allulose 0.4. What is documented is the exposure: mean SGFE intake up to 1.6 mg/kg bw/day, 90th percentile 4.2 mg/kg bw/day in healthy adults.
Net carbs	Subtracts in full — none of its grams countUS FDA, Center for Food Safety and Applied Nutrition regulatory checked 30 Aug 2026 Note FDA: allulose "is virtually unmetabolized in the human body", produces "only a negligible increase in glycemic and insulinemic responses", and provides no more than 0.4 kcal/g.	Subtracts in full — none of its grams countUS FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026 Note Applies to the EXTRACT ONLY. Reported mean dietary exposures are 0.58-1.52 mg/kg bw/day of mogroside V depending on subpopulation — milligrams, against grams for a bulk sweetener. For a retail blend, compute the carbohydrate from the bulking agent and treat the monk fruit as zero.
Blood glucose	Negligible; no established GI value — reduces POSTPRANDIAL glucose and insulin (moderate certainty), but no significant effect on HbA1c, fasting glucose/insulin, lipids, uric acid or body compositionOsborn L, DuPuis K, Liu S, Della Corte D,… peer reviewed checked 31 Aug 2026 Note UPDATED 2026-08-31 with the field's current best synthesis, previously flagged in this backlog as bot-blocked — it is Open Access and loads cleanly in a real browser; only automated fetchers (WebFetch, server curl) were gated, the same JS bot-check pattern already on file for EUR-Lex and the ADA Standards of Care. Osborn, DuPuis, Liu, Della Corte & Della Corte 2026 (AJCN 123(6):101314) is a PROSPERO-registered (CRD42024567433) systematic review and meta-analysis of 20 controlled human intervention trials (12 allulose, 8 tagatose; 1,033 participants total), searched through 30 April 2025 — broader than either single study previously cited here. ALLULOSE ARM: significantly reduced postprandial glucose (incremental AUC SMD -0.66, 95% CI -0.92 to -0.39, moderate certainty) and postprandial insulin (SMD -1.27, 95% CI -2.14 to -0.40, moderate certainty) — but NO significant pooled effect on HbA1c, fasting glucose, fasting insulin, blood lipids, uric acid, or body composition (very low to moderate certainty across those outcomes). Read the postprandial finding and the null findings together: the acute glucose/insulin-blunting effect this record has documented since Tani 2023 holds up under a larger, more rigorous synthesis, but it does not extend to weight, fasting markers or long-term glycaemic control on the evidence gathered so far — allulose blunts a spike, it has not been shown to improve the numbers that matter for longer-run metabolic health. No third-party funding; the review's only disclosed conflict (SL) is unrelated consulting/editorial work with no rare-sugar-industry tie, a materially cleaner disclosure than tani2023's Kagawa University rare-sugar-group authorship (still flagged on that source below). Tani 2023 (145 participants/group, healthy adults, mechanism-establishing) and Ayesh 2024 (126 people with type 2 diabetes, first T2D-specific signal) remain cited in full below as the specific studies this broader 2026 synthesis draws on and updates, not as duplicated or superseded claims.	No meaningful glucose or insulin spike in healthy adults, acutelyTey SL, Salleh NB, Henry J, Forde CG; International… peer reviewed checked 31 Aug 2026 Note Tey et al. 2017, the only published human RCT of monk fruit sweetener itself (as opposed to its regulatory toxicology). 30 healthy men, randomised crossover, four 65 g-sucrose-equivalent preloads (aspartame, monk fruit, stevia, sucrose). Monk fruit, stevia and aspartame all produced significantly lower postprandial glucose and insulin than sucrose, with no meaningful spike above baseline for any of the three. THE MORE INTERESTING FINDING: ad libitum lunch intake after the non-nutritive-sweetener preloads was significantly HIGHER than after sucrose — compensatory eating largely offset the calories saved at the preload, leaving total daily energy intake unchanged across all four arms. Read as evidence about the preload only, not as evidence the sweetener itself changes appetite regulation days or weeks out — one meal, one measurement.
Acceptable daily intake	No sourced entry	Temporary ADI 0-10 mg/kg bw/day, as mogroside V (JECFA, 2026)JECFA (FAO/WHO) regulatory checked 27 Oct 2026 Note JECFA, 102nd meeting (2026): "The Committee established a temporary ADI of 0-10 mg/kg bw per day, expressed as mogroside V, based on a NOAEL of 2144 mg/kg bw per day for mogroside V, the highest dose tested in a 52-week rat study and the application of a 200-fold uncertainty factor." In the same evaluation JECFA took a conservative estimate of the highest dietary exposure for children, 38 mg/kg bw/day, and concluded it "exceeds the temporary ADI by approximately four-fold, which would pose a potential safety concern under the proposed uses and proposed maximum levels". JECFA noted that estimate rests on proposed Codex (GSFA) food categories and maximum levels, not on measured intakes; FDA's modelling for US use (GRN 706, see dietary_exposure_mg_per_kg_day) is far lower, a 1.52 mg/kg bw/day mean for diabetic children. To complete the evaluation JECFA requires, by the end of 2029, a reproductive toxicity study that evaluates the reproductive capacity of the F1 generation, and revised proposed uses and maximum levels. History: JECFA put monk fruit extract on its 2016 agenda and removed it because no data were submitted; EFSA (2019) could not set an ADI because the toxicity database was insufficient (see safety_signal).
Digestive tolerance	0.4 g/kg in one sitting · 0.9 g/kg per day · severe symptoms at 0.5 g/kgHan Y et al., Nutrients 10(12):2010 peer reviewed checked 30 Aug 2026 Note Han 2018, two non-randomised dose-escalation experiments (n=30 and n=19), Korean adults aged 21-30 with BMI 18.5-23. Severe diarrhoea appeared at 0.5 g/kg single dose. The authors explicitly warn that "considerably more symptoms may appear in individuals who are old, obese, or with certain metabolic and GI diseases" and flag the small sample and a possibly inadequate one-week washout. Not a regulatory ADI — no authority has set one.	No sourced entry
Browning	HighXie X, Yu L, Lin Q, Huang D; Journal… peer reviewed checked 30 Aug 2026 Note Browns MORE than sucrose, not less. In sponge cake, lightness fell and redness/yellowness rose with allulose level, with the browning index rising sharply above 20% sucrose replacement and antioxidant activity nearly tripling at the highest level. Practical consequence: the window between caramelised and burnt is narrower than with sugar.	No sourced entry
Bulk contribution	BulkingXie X, Yu L, Lin Q, Huang D; Journal… peer reviewed checked 30 Aug 2026 Note Used at sugar-like weights, but above 20% sucrose replacement in sponge cake hardness and chewiness rose significantly, and specific volume fell from 3.36 to 2.96 cm3/g at 50% replacement. A 1:1 swap also under-sweetens at ~70% relative sweetness.	NoneUS FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026 Note High-intensity sweetener used at milligram levels; contributes no volume, structure, browning or mouthfeel. Any product that fills a measuring cup is doing so with something else. This is why tabletop "monk fruit" is usually a blend.
Recrystallisation	Low at typical food-product use levels; concentrated syrup crystallises on storageRoquette Freres SA patent checked 31 Aug 2026Declared interest: Roquette is a sweetener and starch-derivative ingredient manufacturer with a direct commercial interest in D-allulose syrup formulations. Note SOURCED 2026-08-31, previously flagged unsourced. Runs OPPOSITE to erythritol's mechanism, not the same direction weaker: erythritol recrystallises because it is relatively LOW-solubility for a bulk sweetener (see that record's own recrystallisation_risk note); allulose is markedly MORE soluble, which several independent manufacturer/patent sources describe as the reason it is less prone to gritty recrystallisation than erythritol at typical use concentrations in finished products such as caramel and frozen desserts. That is NOT the same claim as "allulose never crystallises" — a Roquette Frères patent's own storage-stability data (Example 2) found allulose SYRUP at 77% dry solids crystallising out during 4 weeks' storage at 4-15°C, which is exactly why a whole patent literature exists on deliberately crystallisable vs. non-crystallisable D-allulose syrup formulations. Read the two together: the CONCENTRATED-SYRUP ingredient form is a genuine crystallisation risk manufacturers actively manage; a FINISHED product at typical (much lower) allulose concentrations is not. DECLARED INTEREST: Roquette Frères is a sweetener/ingredient manufacturer with a commercial stake in allulose syrup formulations — a different company to the Kagawa/Matsutani interest disclosed on cooling_effect above, which is useful independent corroboration on the general solubility-driven mechanism even though both are industry sources.	No sourced entry
United States	GRAS by notificationUS FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026 Note Multiple "no questions" letters: GRN 400 (18 June 2012), Matsutani (12 June 2014), GRN 693 (Samyang, closed 28 August 2017), and later notices. FDA is explicit that such a letter "is not an affirmation that D-psicose is GRAS under 21 CFR 170.35".	GRAS by notificationUS FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026 Note Six "no questions" letters: GRN 301 (15 Jan 2010), 359 (11 Apr 2011), 522 (8 Dec 2014), 556 (5 Nov 2015), 629, and 706 (17 Aug 2017). FDA is explicit that such a letter "is not an affirmation that SGFE is GRAS under 21 CFR 170.35". Intended use excludes infant formula.
European Union	Not authorisedEFSA Panel on Nutrition, Novel Foods and Food Allergens,… regulatory checked 30 Aug 2026 Note READ THE MECHANISM BEFORE REPEATING THIS. EFSA concluded on 6 May 2025 that "the safety of the NF D-allulose cannot be established" — but the reason is that the applicant stopped responding to EFSA after February 2020, leaving gaps on identity, production process, genotoxicity, stability, batch data and human data. This is an abandoned dossier, NOT a finding of harm. Allulose is a novel food under Regulation (EU) 2015/2283 and cannot lawfully be placed on the EU market without authorisation.	Not authorised as a food; only decoctions are exempt from needing authorisationEuropean Commission, Directorate-General for Health and Food Safety —… regulatory checked 31 Aug 2026 Note CORRECTED 2026-08-31 by two independent sessions running concurrently, reconciled here after a file-write race. This record previously said "Partially authorised", citing Commission Implementing Regulation (EU) 2024/2345 of 7 October 2024 as having authorised an aqueous monk fruit extract as a novel food, taken from Kaim 2025 (a peer-reviewed review). THAT IS WRONG, and it was wrong in the paper it came from. WHAT IS ACTUALLY TRUE, per the European Commission Novel Food Catalogue entry for Siraitia grosvenorii (updated 3 October 2024, source eucatalogue): the FRUIT and its extracts are classified NOVEL FOOD — "not consumed in the EU to a significant degree as a food before 15 May 1997", so "a pre-market authorisation in accordance with Regulation (EU) 2015/2283 is required before it can be placed as food on the EU market" — and no such authorisation has been granted. SEPARATELY, monk fruit DECOCTIONS are classified NOT NOVEL, following a determination by the Irish competent authority (FSAI) under Commission Implementing Regulation (EU) 2018/456, on the basis that decoctions were consumed in the Union before 15 May 1997. A not-novel determination means authorisation is NOT REQUIRED for decoctions specifically; it is not itself an authorisation, and it does not cover purified mogroside V or solvent extracts. Likely origin of the error: the decoction not-novel determination was catalogued in early October 2024 and got telescoped, somewhere in the citation chain, into an "authorising regulation of 7 October 2024" for the extract. Independently corroborated five ways: (1) the ELI lookup for reg_impl:2024:2345:oj returns "No results found"; (2) document number 2024/2345 does resolve on EUR-Lex, but to an unrelated OJ C-series item — a European Investment Bank Deputy Chief Executive vacancy notice published 21 March 2024 — with zero mention of Siraitia, monk fruit or novel food, which is why CELEX 32024R2345 (R for Regulation) 404s; (3) an EUR-Lex advanced search of every Commission Implementing Regulation published in October 2024 returns zero matches for "Siraitia" (the two novel-food authorisations that month were vitamin D2 mushroom powder and magnesium L-threonate); (4) the consolidated Union list of novel foods (Reg (EU) 2017/2470) contains zero occurrences of "Siraitia", "grosvenorii", "monk fruit" or "mogroside" — checked by full-text search of the entire consolidated Annex across three separate consolidated versions (2024-11-07, 2026-07-02, 2026-08-10 — see unionlist2026), with presence checks on known unrelated entries (astaxanthin, krill, chia) confirming each render was complete, not truncated; (5) kaim2025's own bibliography entry for the regulation cites "Off. J. Eur. Union. 2024;L 279:25–28", and that citation also fails to resolve on EUR-Lex ("the requested document does not exist") — the paper cites no distinct EFSA reference for the "2024 EFSA opinion" it says underpins the authorisation either; its footnotes for that claim point only to the 2019 additive opinion and an FSAI page about decoction status. Practical position: monk fruit EXTRACT cannot lawfully be placed on the EU market as a food — neither via the additive route (EFSA found the 2019 toxicology dossier insufficient; no ADI) nor via the novel-food route (no authorisation exists). A DECOCTION is a different product with a different, more permissive answer. The claim of a further additive application accepted for EFSA assessment with public consultation opened April 2026 is retained from kaim2025 but was NOT independently re-verified this pass — treat it as less solid than the findings above, which are.

Two Different Jobs

Allulose is a bulk sweetener. It is used at sugar-like weights and adds body to a recipe, though a one-for-one swap by weight under-sweetens, because it delivers about 70% of sucrose’s sweetness. That figure is widely repeated rather than precisely measured: no one has mapped how it changes with concentration, as has been done for erythritol.

Monk fruit extract does the opposite job. FDA puts the extract at 100 to 250 times sweeter than sugar, depending on how much mogroside V it contains, and that varies enormously: one GRAS notice alone lists nine grades, from 2% to 95% mogroside V, and the label almost never says which grade is in the tub. Used at milligram levels, the extract contributes no volume, structure, browning or mouthfeel. Anything that fills a measuring cup is doing so with a bulking agent, very often erythritol. Judge that product by its bulking agent, because that is where its carbohydrate, its digestive ceiling and its cooling effect come from.

Baking

Allulose is the more sugar-like of the two in the oven, and in one respect more than sugar: it browns more than sucrose. In a 2024 sponge-cake study, browning rose sharply once allulose replaced more than 20% of the sugar, so the window between caramelized and burnt is narrower than bakers are used to. The same study found hardness and chewiness rising above 20% replacement, and specific volume falling from 3.36 to 2.96 cm³/g at 50% replacement. Its high solubility makes it less prone than erythritol to gritty recrystallization in finished foods at typical use levels, though concentrated allulose syrup can crystallize in storage.

Monk fruit extract on its own does nothing structural in a recipe. How a monk fruit product bakes is decided by whatever it is blended with.

Calories and Blood Sugar

US labels may count allulose at 0.4 kcal/g, but only because FDA has said it will not enforce the regulation that still requires 4 kcal/g for carbohydrates while it works on a rule. The same enforcement discretion keeps allulose out of Total and Added Sugars. It still counts in Total Carbohydrate, and it still has to appear in the ingredient list.

No authority has assigned monk fruit extract a caloric factor. What is documented is how little of it people eat: FDA’s exposure modeling found mean mogroside V intakes between 0.58 and 1.52 mg per kg of body weight a day, depending on the group. Its contribution is negligible at those levels, which is a different statement from “zero calories”.

On blood sugar, allulose has the larger evidence base. A 2026 meta-analysis of controlled human trials found it lowered the rise in glucose and insulin after a meal, with moderate certainty, but found no effect on HbA1c, fasting glucose, fasting insulin, blood lipids or body composition. It blunts a spike; it has not been shown to improve the longer-run numbers. A separate 2024 meta-analysis of six trials in 126 people with type 2 diabetes found lower glucose after meals and less time above range on continuous glucose monitors.

Monk fruit sweetener has one published human trial of its own: 30 healthy men, one drink each, where monk fruit, like stevia and aspartame, caused no meaningful glucose or insulin spike. The same study found people ate more at lunch afterward, which offset the calories saved. No trial has tested it in people with diabetes, although FDA’s modeling found children with diabetes have the highest average intake of any group it looked at.

Tolerance and Safety

Allulose has a digestive ceiling, and it is lower than many people assume. In a small dose-escalation study of young Korean adults, symptoms rose above single doses of about 0.4 g per kg of body weight (around 28 g for a 70 kg adult), and severe diarrhea appeared at 0.5 g/kg. The authors warn that older people, people with obesity and people with gastrointestinal conditions may react at lower doses. No authority has set an intake limit for allulose.

Monk fruit’s open question is about toxicology, not digestion. In 2019 EFSA found the toxicity data insufficient to judge its safety as a food additive, and noted a 90-day rat study with decreased testis weight that “cannot be ruled out” as adverse; that is an incomplete file, not a demonstrated harm. In 2026 JECFA set monk fruit’s first acceptable daily intake, and only a temporary one: 0–10 mg per kg of body weight a day, as mogroside V. In the same evaluation JECFA estimated children’s exposure under proposed international uses at about four times that figure, which it called a potential safety concern. That estimate rests on proposed uses, not measured intakes, and FDA’s modeling of US use is far lower. JECFA wants a reproductive toxicity study and revised proposed uses by the end of 2029 to finish its evaluation.

Regulatory Status

In the US, both are GRAS by notification: FDA has issued “no questions” letters for allulose since 2012 and six for monk fruit extracts between 2010 and 2017. A “no questions” letter is not a formal FDA approval; it means FDA did not object to the notifier’s own conclusion.

In the EU, neither is authorized, and the reasons differ. Allulose’s novel-food application stalled when the applicant stopped answering EFSA’s questions after February 2020; EFSA concluded in 2025 that safety “cannot be established”, which is an abandoned dossier rather than a finding of harm. Monk fruit extract is classed as a novel food with no authorization, and EFSA’s 2019 additive opinion found the data insufficient. Only monk fruit decoctions, a traditional preparation, are exempt from needing authorization; that does not cover the purified extract.

Best Use Cases

Reach for allulose when the sweetener has to do sugar’s structural work: browning, body, a soft crumb, a sauce that stays smooth. Use it below the digestive ceiling, and lower the oven heat or watch the color, because it browns faster than sugar.

Reach for monk fruit extract when you only need sweetness: drinks, dressings, anything where a few milligrams replace a spoonful. Read the ingredient list of any cup-for-cup product, because its bulking agent is what you are mostly eating.

A blend is common for good reason: allulose supplies the bulk and browning, and a little monk fruit makes up the 30% sweetness allulose lacks.

The Bottom Line

Allulose and monk fruit are complements more than rivals. Allulose is the closer functional match to sugar, with real evidence that it blunts after-meal glucose and a digestive limit worth respecting. Monk fruit extract adds sweetness with nothing else attached, but its safety file is the thinner of the two: its only ADI is temporary, and the evaluation behind it is not finished. Neither is on the EU market, and in the US both rest on GRAS notifications rather than formal approvals.

// Sourced record

Allulose — the checked figures

Supported but limited Reviewed 31 Aug 2026 Next review 28 Feb 2027

A sugar substitute cannot treat a low blood glucose

Low blood glucose is below 70 mg/dL. The ADA's stated rule is 15 grams of FAST-ACTING carbohydrate, then recheck after 15 minutes — none of the substitutes on this site qualify, because not raising blood glucose is the entire point of them. Severe hypoglycaemia is an emergency needing help from others; if someone is unconscious and glucagon is unavailable or nobody knows how to use it, call 911. Your own care plan takes priority over anything written here.

American Diabetes Association clinical guidance checked 31 Aug 2026

Sweetness: 70% of sugar Commonly cited as approximately 70% of sucrose. Convergent across Xie 2024, Han 2018 and the comments FDA received on the 2014 proposed rule — but this is a repeated figure, not a specification band or a sensory measurement, and no concentration dependence has been characterised the way it has for erythritol. Xie X, Yu L, Lin Q, Huang D; Journal… peer reviewed checked 30 Aug 2026
Calories: 0.4 kcal per gram ENFORCEMENT DISCRETION, not a rule. FDA states that under 21 CFR 101.9(c)(1)(B) "a caloric value of 4 kcal/g must currently be used for allulose because it is a carbohydrate", and that it merely intends not to enforce that pending rulemaking. Two citizen petitions asked for 0.4 and 0.2; FDA chose the higher figure because it "would not underestimate the caloric contribution". US FDA, Center for Food Safety and Applied Nutrition regulatory checked 30 Aug 2026
Net carbs: Subtracts in full — none of its grams count FDA: allulose "is virtually unmetabolized in the human body", produces "only a negligible increase in glycemic and insulinemic responses", and provides no more than 0.4 kcal/g. US FDA, Center for Food Safety and Applied Nutrition regulatory checked 30 Aug 2026 "Net carbs" is not a recognised measure — start from Total Carbohydrate The American Diabetes Association is explicit: the term "does not have a legal definition and is not used by the Food and Drug Administration or recognized by American Diabetes Association", and the FDA recommends using total carbohydrates on the label. Their reasoning applies directly to the subtraction this site performs: it assumes fibre and sugar alcohols are not absorbed or metabolised, "but this is not always true, and some are partially digested and therefore still provide calories as well as impact blood glucose" — and because the label does not say WHICH sugar alcohol is present, the effect on blood glucose "cannot be determined precisely". Treat the figure above as a formulation aid, not a carbohydrate count to dose insulin against. American Diabetes Association clinical guidance checked 31 Aug 2026
Glycemic response: negligible; no established GI value — reduces POSTPRANDIAL glucose and insulin (moderate certainty), but no significant effect on HbA1c, fasting glucose/insulin, lipids, uric acid or body composition UPDATED 2026-08-31 with the field's current best synthesis, previously flagged in this backlog as bot-blocked — it is Open Access and loads cleanly in a real browser; only automated fetchers (WebFetch, server curl) were gated, the same JS bot-check pattern already on file for EUR-Lex and the ADA Standards of Care. Osborn, DuPuis, Liu, Della Corte & Della Corte 2026 (AJCN 123(6):101314) is a PROSPERO-registered (CRD42024567433) systematic review and meta-analysis of 20 controlled human intervention trials (12 allulose, 8 tagatose; 1,033 participants total), searched through 30 April 2025 — broader than either single study previously cited here. ALLULOSE ARM: significantly reduced postprandial glucose (incremental AUC SMD -0.66, 95% CI -0.92 to -0.39, moderate certainty) and postprandial insulin (SMD -1.27, 95% CI -2.14 to -0.40, moderate certainty) — but NO significant pooled effect on HbA1c, fasting glucose, fasting insulin, blood lipids, uric acid, or body composition (very low to moderate certainty across those outcomes). Read the postprandial finding and the null findings together: the acute glucose/insulin-blunting effect this record has documented since Tani 2023 holds up under a larger, more rigorous synthesis, but it does not extend to weight, fasting markers or long-term glycaemic control on the evidence gathered so far — allulose blunts a spike, it has not been shown to improve the numbers that matter for longer-run metabolic health. No third-party funding; the review's only disclosed conflict (SL) is unrelated consulting/editorial work with no rare-sugar-industry tie, a materially cleaner disclosure than tani2023's Kagawa University rare-sugar-group authorship (still flagged on that source below). Tani 2023 (145 participants/group, healthy adults, mechanism-establishing) and Ayesh 2024 (126 people with type 2 diabetes, first T2D-specific signal) remain cited in full below as the specific studies this broader 2026 synthesis draws on and updates, not as duplicated or superseded claims. Osborn L, DuPuis K, Liu S, Della Corte D,… peer reviewed checked 31 Aug 2026
Counts in Total Carbohydrate (US): Yes A determination, not discretion. FDA: "allulose must be included in the amount of Total Carbohydrate declared on the label under the existing regulations." Carbohydrate classification is by chemical structure, not physiological effect. US FDA, Center for Food Safety and Applied Nutrition regulatory checked 30 Aug 2026
Counts in Total Sugars (US): No Excluded by enforcement discretion pending rulemaking — the regulation itself still captures allulose as a free monosaccharide. Grounds: not cariogenic (plaque pH does not fall below 5.7), much less than 4 kcal/g, negligible glycaemic and insulinaemic response. US FDA, Center for Food Safety and Applied Nutrition regulatory checked 30 Aug 2026
Counts in Added Sugars (US): No Follows Total Sugars, of which Added Sugars is a subset. Excluded from both the gram amount and the %DV. Enforcement discretion. US FDA, Center for Food Safety and Applied Nutrition regulatory checked 30 Aug 2026
Ingredient list: Yes FDA notes this twice: excluding allulose from the sugar lines does not exclude it from the ingredient statement under 21 CFR 101.4. US FDA, Center for Food Safety and Applied Nutrition regulatory checked 30 Aug 2026
Digestive tolerance: 0.4 g/kg in one sitting · 0.9 g/kg per day · severe symptoms at 0.5 g/kg Han 2018, two non-randomised dose-escalation experiments (n=30 and n=19), Korean adults aged 21-30 with BMI 18.5-23. Severe diarrhoea appeared at 0.5 g/kg single dose. The authors explicitly warn that "considerably more symptoms may appear in individuals who are old, obese, or with certain metabolic and GI diseases" and flag the small sample and a possibly inadequate one-week washout. Not a regulatory ADI — no authority has set one. Han Y et al., Nutrients 10(12):2010 peer reviewed checked 30 Aug 2026
Browning: high Browns MORE than sucrose, not less. In sponge cake, lightness fell and redness/yellowness rose with allulose level, with the browning index rising sharply above 20% sucrose replacement and antioxidant activity nearly tripling at the highest level. Practical consequence: the window between caramelised and burnt is narrower than with sugar. Xie X, Yu L, Lin Q, Huang D; Journal… peer reviewed checked 30 Aug 2026
Bulk contribution: bulking Used at sugar-like weights, but above 20% sucrose replacement in sponge cake hardness and chewiness rose significantly, and specific volume fell from 3.36 to 2.96 cm3/g at 50% replacement. A 1:1 swap also under-sweetens at ~70% relative sweetness. Xie X, Yu L, Lin Q, Huang D; Journal… peer reviewed checked 30 Aug 2026
Recrystallisation: Low at typical food-product use levels; concentrated syrup crystallises on storage SOURCED 2026-08-31, previously flagged unsourced. Runs OPPOSITE to erythritol's mechanism, not the same direction weaker: erythritol recrystallises because it is relatively LOW-solubility for a bulk sweetener (see that record's own recrystallisation_risk note); allulose is markedly MORE soluble, which several independent manufacturer/patent sources describe as the reason it is less prone to gritty recrystallisation than erythritol at typical use concentrations in finished products such as caramel and frozen desserts. That is NOT the same claim as "allulose never crystallises" — a Roquette Frères patent's own storage-stability data (Example 2) found allulose SYRUP at 77% dry solids crystallising out during 4 weeks' storage at 4-15°C, which is exactly why a whole patent literature exists on deliberately crystallisable vs. non-crystallisable D-allulose syrup formulations. Read the two together: the CONCENTRATED-SYRUP ingredient form is a genuine crystallisation risk manufacturers actively manage; a FINISHED product at typical (much lower) allulose concentrations is not. DECLARED INTEREST: Roquette Frères is a sweetener/ingredient manufacturer with a commercial stake in allulose syrup formulations — a different company to the Kagawa/Matsutani interest disclosed on cooling_effect above, which is useful independent corroboration on the general solubility-driven mechanism even though both are industry sources. Roquette Freres SA patent checked 31 Aug 2026Declared interest: Roquette is a sweetener and starch-derivative ingredient manufacturer with a direct commercial interest in D-allulose syrup formulations.
Cooling effect: Present, comparable in magnitude to erythritol SOURCED 2026-08-31, previously flagged unsourced. A direct sensory/calorimetric measurement in a Matsutani Chemical/Kagawa University patent found allulose's endothermic heat of dissolution at 27.4 cal/g (4936 cal/mol, approx. 20.7 kJ/mol) — the patent's own text calls this "almost similar" to erythritol, which it separately measured at 5239 cal/mol (approx. 21.9 kJ/mol) in the same experiment. That erythritol figure lines up closely with this file's own independently-sourced erythritol value (22-24 kJ/mol, tyapkova2012, a different paper entirely) — real cross-validation, not the same number repeated. DECLARED INTEREST: co-owned by Kagawa University's rare-sugar group, the same authorship line already flagged on tani2023 elsewhere in this record; it is nonetheless the only direct calorimetric measurement of allulose's cooling effect found this pass, patent examples are original experimental data rather than marketing claims, and the erythritol cross-check argues the method is sound. Matsutani Chemical Industry Co., Ltd. and Kagawa University NUC… patent checked 31 Aug 2026Declared interest: Co-assignee Kagawa University is the same rare-sugar research group whose authorship is already flagged as a conflict on tani2023 in this record. A patent example is original experimental data, not a peer-reviewed independent replication.
United States: GRAS by notification Multiple "no questions" letters: GRN 400 (18 June 2012), Matsutani (12 June 2014), GRN 693 (Samyang, closed 28 August 2017), and later notices. FDA is explicit that such a letter "is not an affirmation that D-psicose is GRAS under 21 CFR 170.35". US FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026
European Union: Not authorised READ THE MECHANISM BEFORE REPEATING THIS. EFSA concluded on 6 May 2025 that "the safety of the NF D-allulose cannot be established" — but the reason is that the applicant stopped responding to EFSA after February 2020, leaving gaps on identity, production process, genotoxicity, stability, batch data and human data. This is an abandoned dossier, NOT a finding of harm. Allulose is a novel food under Regulation (EU) 2015/2283 and cannot lawfully be placed on the EU market without authorisation. EFSA Panel on Nutrition, Novel Foods and Food Allergens,… regulatory checked 30 Aug 2026

Where this leaves it

The most sugar-like behaviour of any low-calorie sweetener covered here, on the least settled regulatory footing. It browns, it does not spike glucose, and it may blunt the glucose response to other carbohydrates — but its US label advantage is enforcement discretion rather than law, and it is not authorised in the EU at all.

Its digestive ceiling is LOWER than erythritol's, not higher: 0.4 g/kg body weight in a single sitting, about 28 g for a 70 kg adult, with severe diarrhoea at 0.5 g/kg.

If you're managing diabetes

The only sweetener in this file with its own clinical trial data in people with type 2 diabetes — a real signal, still small, and the digestive ceiling is tighter than erythritol's.

Ayesh et al. 2024 (6 RCTs, 126 people with diagnosed type 2 diabetes, 5-8.5 g per meal) found a significant reduction in glucose AUC and in continuous-glucose-monitor time-above-range; fasting glucose and time-in-range did not reach significance. That sits alongside Tani 2023's larger healthy-adult AUC-reduction finding, not in place of it — see glycemic_response above. Net carbs subtract in full and the glycaemic/insulinaemic response is negligible per FDA. The caution: allulose's single-dose tolerance ceiling (0.4 g/kg bw, about 28 g for a 70 kg adult) is LOWER than erythritol's, not higher, and severe diarrhoea appeared at 0.5 g/kg in the dosing study — worth knowing before treating it as the gentler option at typical sugar-replacement quantities.

Ayesh H, Suhail S, Ayesh S; Metabolism Open 24:100329 peer reviewed checked 31 Aug 2026

What would change this: FDA completing the rulemaking that would convert the Total Sugars exclusion and the 0.4 kcal/g factor from enforcement discretion into regulation, or a new EU novel-food application from an applicant willing to answer EFSA's data requests.

No sweetener here is established to improve long-term health ADA Standards of Care 2026, Section 5: "NNS do not appear to have a significant effect on glycemic management," and evidence on weight is mixed — some reviews find modest weight-loss benefit, others an association with weight gain, likely reflecting reverse causation and confounding; added to an eating plan without energy restriction, NNS show no weight benefit. Substituting can reduce sugar and calories. That is a different claim from improving an outcome, and only the first is supported. CORROBORATED 2026-08-31 at much larger scale: Jeong et al.'s umbrella review (29 meta-analyses, 24 observational studies plus 6 RCTs, 50M+ participants, 135 associations examined) found observational studies link higher sweetener intake to INCREASED risk of weight gain, metabolic syndrome, type 2 diabetes and hypertension, while RCTs show only modest weight-loss benefit versus sugar specifically. The two study designs are not measuring the same comparison — see the general-health page for why — but neither supports a durable, independent long-term health benefit, which is the ADA position this guardrail already states. American Diabetes Association Professional Practice Committee for Diabetes; Diabetes… clinical guidance checked 31 Aug 2026

Every figure above links to the document it came from and carries the date we last opened that document. Where a figure has no source it is not shown at all, rather than stated without one. Tell us if you find one that is wrong.

// Sourced record

Monk Fruit — the checked figures

Preliminary Reviewed 27 Oct 2026 Next review 31 Dec 2026

A sugar substitute cannot treat a low blood glucose

American Diabetes Association clinical guidance checked 31 Aug 2026

Sweetness: 100–250× sweeter than sugar A MULTIPLE, not a fraction — see the header warning. CORRECTED 2026-08-30 from an earlier 100-400 band: FDA gives 100-250 for the EXTRACT ("SGFE, depending on the mogroside content, is reported to be 100 to 250 times sweeter than sugar"), while the ~250 figure in the literature describes PURE mogroside V. Those are not in conflict once you separate extract from active compound, and the extract figure is the one that matters commercially. Potency tracks mogroside V content, which varies 2%-95% between grades. US Food and Drug Administration regulatory checked 30 Aug 2026
Active compound content: 2–95% mogroside V, depending on grade Nine distinct product grades in a single GRAS notice (GRN 629): 2%, 18-23%, 24-27%, 40%, 45%, 50%, 60%, 90%, 95% mogroside V, separated by successive resin and chromatography steps. Across GRNs 301/359/522/556 the range is 12.5-90%. The label almost never states which grade is in the tub. US FDA GRAS Notice Inventory regulatory checked 30 Aug 2026
Calories: negligible at use levels Deliberately not stated as "0 kcal/g". No authority has assigned monk fruit extract a caloric factor the way FDA assigned erythritol 0 and allulose 0.4. What is documented is the exposure: mean SGFE intake up to 1.6 mg/kg bw/day, 90th percentile 4.2 mg/kg bw/day in healthy adults. US FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026
Net carbs: Subtracts in full — none of its grams count Applies to the EXTRACT ONLY. Reported mean dietary exposures are 0.58-1.52 mg/kg bw/day of mogroside V depending on subpopulation — milligrams, against grams for a bulk sweetener. For a retail blend, compute the carbohydrate from the bulking agent and treat the monk fruit as zero. US FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026 "Net carbs" is not a recognised measure — start from Total Carbohydrate The American Diabetes Association is explicit: the term "does not have a legal definition and is not used by the Food and Drug Administration or recognized by American Diabetes Association", and the FDA recommends using total carbohydrates on the label. Their reasoning applies directly to the subtraction this site performs: it assumes fibre and sugar alcohols are not absorbed or metabolised, "but this is not always true, and some are partially digested and therefore still provide calories as well as impact blood glucose" — and because the label does not say WHICH sugar alcohol is present, the effect on blood glucose "cannot be determined precisely". Treat the figure above as a formulation aid, not a carbohydrate count to dose insulin against. American Diabetes Association clinical guidance checked 31 Aug 2026
Glycemic response: No meaningful glucose or insulin spike in healthy adults, acutely Tey et al. 2017, the only published human RCT of monk fruit sweetener itself (as opposed to its regulatory toxicology). 30 healthy men, randomised crossover, four 65 g-sucrose-equivalent preloads (aspartame, monk fruit, stevia, sucrose). Monk fruit, stevia and aspartame all produced significantly lower postprandial glucose and insulin than sucrose, with no meaningful spike above baseline for any of the three. THE MORE INTERESTING FINDING: ad libitum lunch intake after the non-nutritive-sweetener preloads was significantly HIGHER than after sucrose — compensatory eating largely offset the calories saved at the preload, leaving total daily energy intake unchanged across all four arms. Read as evidence about the preload only, not as evidence the sweetener itself changes appetite regulation days or weeks out — one meal, one measurement. Tey SL, Salleh NB, Henry J, Forde CG; International… peer reviewed checked 31 Aug 2026
Acceptable daily intake: Temporary ADI 0-10 mg/kg bw/day, as mogroside V (JECFA, 2026) JECFA, 102nd meeting (2026): "The Committee established a temporary ADI of 0-10 mg/kg bw per day, expressed as mogroside V, based on a NOAEL of 2144 mg/kg bw per day for mogroside V, the highest dose tested in a 52-week rat study and the application of a 200-fold uncertainty factor." In the same evaluation JECFA took a conservative estimate of the highest dietary exposure for children, 38 mg/kg bw/day, and concluded it "exceeds the temporary ADI by approximately four-fold, which would pose a potential safety concern under the proposed uses and proposed maximum levels". JECFA noted that estimate rests on proposed Codex (GSFA) food categories and maximum levels, not on measured intakes; FDA's modelling for US use (GRN 706, see dietary_exposure_mg_per_kg_day) is far lower, a 1.52 mg/kg bw/day mean for diabetic children. To complete the evaluation JECFA requires, by the end of 2029, a reproductive toxicity study that evaluates the reproductive capacity of the F1 generation, and revised proposed uses and maximum levels. History: JECFA put monk fruit extract on its 2016 agenda and removed it because no data were submitted; EFSA (2019) could not set an ADI because the toxicity database was insufficient (see safety_signal). JECFA (FAO/WHO) regulatory checked 27 Oct 2026
Measured intake: healthy adults 0.58 · diabetic adults 0.63 · healthy children 0.96 · diabetic children 1.52 mg/kg per day (mean) Mogroside V, per GRN 706. 90th-percentile figures are 1.53 / 2.03 / 2.24 / 2.05 for the same four groups. Diabetic children carry the highest mean exposure. US FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026
Open safety question: Unresolved testicular finding in a 90-day rat study; toxicity database judged insufficient EFSA 2019: "the toxicity database on Monk fruit extract is insufficient to conclude on the safety of the use of monk fruit extract as a food additive." Specific gaps: no chronic/carcinogenicity study despite mogroside V being systemically bioavailable; in vitro genotoxicity insufficient to cover microbial-degradation metabolites including the aglycone; male exposure in the reproductive screening study "did not cover the full length of spermatogenesis"; and a 90-day rat study showed decreased testis weight with histopathological findings that "cannot be ruled out" as adverse. This is an incomplete dossier, not a demonstrated harm. EFSA Panel on Food Additives and Flavourings, EFSA Journal… regulatory checked 30 Aug 2026
Bulk contribution: none High-intensity sweetener used at milligram levels; contributes no volume, structure, browning or mouthfeel. Any product that fills a measuring cup is doing so with something else. This is why tabletop "monk fruit" is usually a blend. US FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026
United States: GRAS by notification Six "no questions" letters: GRN 301 (15 Jan 2010), 359 (11 Apr 2011), 522 (8 Dec 2014), 556 (5 Nov 2015), 629, and 706 (17 Aug 2017). FDA is explicit that such a letter "is not an affirmation that SGFE is GRAS under 21 CFR 170.35". Intended use excludes infant formula. US FDA, Office of Food Additive Safety regulatory checked 30 Aug 2026
European Union: Not authorised as a food; only decoctions are exempt from needing authorisation CORRECTED 2026-08-31 by two independent sessions running concurrently, reconciled here after a file-write race. This record previously said "Partially authorised", citing Commission Implementing Regulation (EU) 2024/2345 of 7 October 2024 as having authorised an aqueous monk fruit extract as a novel food, taken from Kaim 2025 (a peer-reviewed review). THAT IS WRONG, and it was wrong in the paper it came from. WHAT IS ACTUALLY TRUE, per the European Commission Novel Food Catalogue entry for Siraitia grosvenorii (updated 3 October 2024, source eucatalogue): the FRUIT and its extracts are classified NOVEL FOOD — "not consumed in the EU to a significant degree as a food before 15 May 1997", so "a pre-market authorisation in accordance with Regulation (EU) 2015/2283 is required before it can be placed as food on the EU market" — and no such authorisation has been granted. SEPARATELY, monk fruit DECOCTIONS are classified NOT NOVEL, following a determination by the Irish competent authority (FSAI) under Commission Implementing Regulation (EU) 2018/456, on the basis that decoctions were consumed in the Union before 15 May 1997. A not-novel determination means authorisation is NOT REQUIRED for decoctions specifically; it is not itself an authorisation, and it does not cover purified mogroside V or solvent extracts. Likely origin of the error: the decoction not-novel determination was catalogued in early October 2024 and got telescoped, somewhere in the citation chain, into an "authorising regulation of 7 October 2024" for the extract. Independently corroborated five ways: (1) the ELI lookup for reg_impl:2024:2345:oj returns "No results found"; (2) document number 2024/2345 does resolve on EUR-Lex, but to an unrelated OJ C-series item — a European Investment Bank Deputy Chief Executive vacancy notice published 21 March 2024 — with zero mention of Siraitia, monk fruit or novel food, which is why CELEX 32024R2345 (R for Regulation) 404s; (3) an EUR-Lex advanced search of every Commission Implementing Regulation published in October 2024 returns zero matches for "Siraitia" (the two novel-food authorisations that month were vitamin D2 mushroom powder and magnesium L-threonate); (4) the consolidated Union list of novel foods (Reg (EU) 2017/2470) contains zero occurrences of "Siraitia", "grosvenorii", "monk fruit" or "mogroside" — checked by full-text search of the entire consolidated Annex across three separate consolidated versions (2024-11-07, 2026-07-02, 2026-08-10 — see unionlist2026), with presence checks on known unrelated entries (astaxanthin, krill, chia) confirming each render was complete, not truncated; (5) kaim2025's own bibliography entry for the regulation cites "Off. J. Eur. Union. 2024;L 279:25–28", and that citation also fails to resolve on EUR-Lex ("the requested document does not exist") — the paper cites no distinct EFSA reference for the "2024 EFSA opinion" it says underpins the authorisation either; its footnotes for that claim point only to the 2019 additive opinion and an FSAI page about decoction status. Practical position: monk fruit EXTRACT cannot lawfully be placed on the EU market as a food — neither via the additive route (EFSA found the 2019 toxicology dossier insufficient; no ADI) nor via the novel-food route (no authorisation exists). A DECOCTION is a different product with a different, more permissive answer. The claim of a further additive application accepted for EFSA assessment with public consultation opened April 2026 is retained from kaim2025 but was NOT independently re-verified this pass — treat it as less solid than the findings above, which are. European Commission, Directorate-General for Health and Food Safety —… regulatory checked 31 Aug 2026

Where this leaves it

The cleanest label story of the three and the thinnest safety file. It is plant-derived, used in milligrams, and carries no caloric or glycaemic load worth counting — but its only ADI is a temporary one JECFA set in 2026, which JECFA's own conservative estimate of children's exposure under proposed Codex uses exceeds about four-fold; no EU authorisation of any monk fruit extract could be confirmed against primary law despite a widely-repeated claim to the contrary (see eu_status); and an unresolved testicular finding sits in the record because the toxicology dossier is still incomplete.

What you buy in a tub labelled "monk fruit" is usually mostly a bulking agent by weight. Judge that product by its bulking agent — very often erythritol — because that is where its carbohydrate, its digestive ceiling and its cooling effect actually come from.

If you're managing diabetes

No published trial has tested monk fruit sweetener in people with diabetes specifically — despite diabetic children showing the highest modelled exposure of any group FDA looked at.

FDA's own exposure modelling in GRN 706 found diabetic children carry the highest mean mogroside V intake of any subgroup assessed (1.52 mg/kg bw/day, against 0.58 for healthy adults) — the population already consuming it most is exactly the one with no dedicated clinical trial behind it. The only human RCT of the sweetener itself (Tey 2017) tested 30 healthy men, not people with diabetes, and found no meaningful glucose or insulin spike from monk fruit, stevia or aspartame versus sucrose. Two things this record already flags matter more here than usual: the only ADI is JECFA's temporary 2026 one, which JECFA's conservative estimate of children's exposure under proposed Codex uses exceeds about four-fold (see adi_status), and what you buy in a tub is usually mostly a bulking agent by weight, very often erythritol — read that record's diabetes note for what you're actually eating in bulk.

US FDA, Office of Food Additive Safety regulatory checked 31 Aug 2026

What would change this: A genuine novel-food application for the extract being filed and reaching a decision, or the EFSA additive assessment reportedly opened to consultation in April 2026 (itself unverified this pass) reaching a conclusion; or the reproductive toxicity study (F1 reproductive capacity) and revised proposed uses that JECFA requires by the end of 2029 to complete its 2026 evaluation and decide whether its temporary ADI stands.

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.