Reb A vs Reb M: Taste, Production & Formulation Differences

Written and reviewed by Dr. Xiaoyan Qiu · August 2026

Rebaudioside A and Rebaudioside M are both steviol glycosides extracted or derived from Stevia rebaudiana Bertoni, but they occupy very different positions in the sweetener supply chain. Reb A is the most abundant steviol glycoside in the stevia leaf, accounting for 2–4% of dry leaf weight, and has been the workhorse of the commercial stevia industry for over a decade. Reb M, by contrast, exists at trace levels in the leaf (below 0.1% dry weight) yet delivers a sweetness profile closer to sucrose than any other known steviol glycoside.^[1][2]

For B2B formulators, the decision between these two grades is not simply a matter of taste preference — it involves trade-offs across production cost, supply-chain transparency, regulatory pathway, and application-specific performance. This article compares Reb A and Reb M on every axis that matters for product development, from the molecular mechanism behind stevia’s aftertaste to the enzymatic bioconversion process that makes commercial-scale Reb M feasible.

What Are Reb A and Reb M?

Both compounds share the same steviol aglycone backbone. The difference lies in the number and arrangement of glucose units attached to that backbone. Reb A carries four glucose moieties, while Reb M carries six — three on the C-13 position and three on the C-19 position, making it one of the most heavily glycosylated members of the steviol glycoside family.^[2]

This structural difference has a direct sensory consequence. The human bitter taste system relies on approximately 25 hTAS2R receptors, and two of them — hTAS2R4 and hTAS2R14 — are responsible for detecting the bitter off-taste of steviol glycosides. Hellfritsch et al. (2012) demonstrated that longer glycone chains physically block access to these bitter receptors, which explains why Reb M (six glucose units) activates them far less than stevioside (two glucose units) or Reb A (four glucose units).^[1] In practical terms, Reb A delivers a sweetness intensity of approximately 200–300× sucrose but carries a recognizable bitter-licorice tail at higher concentrations. Reb M delivers 200–350× sucrose sweetness with a rounder onset, fuller mouthfeel, and minimal lingering bitterness.^[2][3]

For a detailed explanation of how stevioside and Reb A differ structurally from the full stevia glycoside family, see our companion articles.

Reb A vs Reb M: Side-by-Side Comparison

The table below summarizes the key differences that matter most for sourcing and formulation decisions. The most significant gap between the two grades is not sweetness intensity — they overlap considerably in that range — but rather the combination of taste quality and unit cost. Reb M typically commands a 3–5× price premium over Reb A at equivalent purity, which is why the formulation decision often comes down to whether the application matrix can tolerate Reb A’s aftertaste or whether the end product demands Reb M’s sugar-like profile.

Feature	Reb A	Reb M
Sweetness multiplier	~200–300× sucrose	~200–350× sucrose
Taste profile	Bitter, licorice aftertaste at high concentration	Sugar-like; virtually no bitterness or aftertaste
Content in stevia leaf	2–4% dry leaf weight	<0.1% dry leaf weight
Primary production method	Direct plant extraction	Enzymatic bioconversion or microbial fermentation
Cost	Lower	Significantly higher
Solubility	Good	Good; superior under certain conditions
Heat stability	Stable	Stable
pH stability	Broad range	Broad range
Regulatory status	FDA GRAS; widely approved globally	FDA GRAS; approved in most major markets
Typical applications	Beverages, baking, condiments, dairy	Premium beverages, health foods, zero-sugar products

Why Stevia Has an Aftertaste — and How Reb M Changes That

The phrase “stevia aftertaste” is one of the most common objections formulators encounter from sensory panels and end-brand customers. Understanding where it comes from — and why it is not an inherent property of all steviol glycosides — is essential for making informed sourcing decisions.

The primary driver of stevia bitterness is stevioside, which carries only two glucose units on its C-13 sophorose chain. This compact structure allows the molecule to enter the binding pocket of hTAS2R4 and hTAS2R14, triggering bitter signal transduction.^[1] Reb A, with an additional C-13 glucose branching at the 3″ position, partially blocks this binding — which is why high-purity Reb A (≥97%) already tastes significantly cleaner than crude stevia leaf extract. However, at concentrations above roughly 5% sucrose equivalence, trained panels consistently detect a lingering licorice note and delayed bitterness return in Reb A solutions.^[3]

Reb M resolves this problem at the molecular level. Its six glucose units create a bulkier glycone cage around the steviol core, effectively preventing meaningful interaction with either bitter receptor. In consumer sensory panels, Reb M at 14% sucrose equivalence showed no statistically significant difference from sucrose in overall sweetness intensity, while Reb A at the same concentration was rated significantly more bitter and astringent.^[3]

One nuance worth noting for formulators: Reb M does exhibit a slightly delayed sweetness onset and a longer sweet extinction time compared to sucrose. Prakash et al. (2014) documented this temporal shift in controlled tasting panels — the sweet signal from Reb M appears approximately 1–2 seconds later and lingers longer than sucrose after swallowing.^[2] This is generally perceived as a positive attribute (a “fuller” sweetness experience), but in applications where an immediate, clean-cutting sweetness is required — such as carbonated soft drinks with very short sip cycles — it is a factor to account for in blend design.

How Reb M Is Produced: Extraction, Enzymatic Bioconversion, and Fermentation

One of the most important distinctions in today’s Reb M supply chain is how the material is actually manufactured. Three production pathways exist, each with different implications for cost, labeling, regulatory classification, and supply-chain transparency.

Traditional Stevia Leaf Extraction

Standard water-ethanol extraction from Stevia rebaudiana leaves yields a glycoside mixture where Reb A dominates at 2–4% of dry leaf weight. Reb M, however, exists at concentrations below 0.1% — making direct extraction and purification to high-purity grades (≥95%) economically impractical at scale.^[4] This route remains viable only for Reb A production and for producing mixed-glycoside stevia extracts where Reb M appears as a minor component.

Enzymatic Bioconversion

This is currently the most commercially relevant pathway for producing high-purity Reb M while retaining a stevia-leaf-derived supply chain. The process begins with purified Reb A or stevioside extracted from stevia leaves, then uses UDP-glucosyltransferase (UGT) enzymes — typically UGT76G1 and UGT91C1 — in combination with a sucrose synthase (SuSy) for continuous UDP-glucose regeneration.^[4]

The reaction proceeds stepwise: Reb A is first glucosylated to Reb D, then Reb D is further glucosylated to Reb M. Because the starting material is stevia leaf extract, the final product retains its “stevia-derived” identity, which is significant for clean-label positioning. The European Food Safety Authority classifies enzymatically bioconverted Reb M under E 960c, and in 2021 the EFSA FAF Panel confirmed there is no safety concern with this production process, maintaining the existing steviol glycoside ADI of 4 mg/kg body weight per day expressed as steviol equivalents.^[5]

Typical commercial output from optimized enzymatic cascades reaches ≥95% Reb M purity by HPLC, with high batch-to-batch consistency. Valeherb’s enzymatically bioconverted Reb M 95% and Reb A 95% are both produced through this pathway.

Microbial Fermentation (De Novo Biosynthesis)

An alternative route bypasses the stevia leaf entirely. Genetically engineered yeast strains — commonly Yarrowia lipolytica or Saccharomyces cerevisiae — are fed simple sugars (cane sugar or corn-derived glucose) and reconstruct the entire steviol glycoside biosynthetic pathway intracellularly, producing Reb M from scratch.^[4]

The resulting molecule is structurally identical to leaf-derived Reb M. However, because the production host is a genetically modified organism and the starting material is not stevia leaf, fermentation-derived Reb M cannot carry Non-GMO, Organic, or “stevia-derived” labeling in most markets. This distinction matters for brands targeting clean-label, non-GMO, or organic positioning.

The table below summarizes the practical differences between these three routes.

Formulation Guide: Which Stevia Grade Fits Your Application?

Choosing between Reb A and Reb M — or deciding on a blend — depends on the application matrix, target consumer positioning, and cost tolerance. The following guide maps common B2B application categories to their best-fit stevia grade, based on how each product matrix interacts with stevia’s sensory profile.

A general principle for formulators: the simpler the flavor matrix, the more important Reb M becomes. In water-based systems with minimal masking ingredients (still water, sparkling water, clear-label functional beverages), any Reb A bitterness is fully exposed. In complex matrices — baked goods with butter, chocolate dairy with cocoa, savory condiments — Reb A performs adequately at a fraction of the cost.

The protein powder and RTD category deserves special attention. Whey protein hydrolysates and plant-based protein isolates (pea, soy, rice) carry their own inherent bitterness from hydrophobic peptide fragments. When Reb A is used as the sweetener in these systems, the bitter notes stack — the protein’s peptide bitterness and Reb A’s glycoside bitterness activate overlapping receptor populations, producing a compounded off-taste that neither ingredient would create alone. Reb M avoids this stacking effect because it does not meaningfully activate the same bitter receptors, making it the preferred choice for protein-based applications even when cost pressure is significant.^[3]

For formulators exploring sweetener blending beyond stevia, monk fruit extract offers complementary temporal characteristics that can round out a stevia-based system. Allulose is another effective pairing, providing bulk and mouthfeel without caloric load. Our zero-calorie sweetener comparison guide covers the full landscape of available options.

Regulatory Status

Both Reb A and Reb M have achieved regulatory clearance in most major markets, though through different pathways. Reb A has the longer regulatory track record, while Reb M’s approval timeline varies depending on whether the material is leaf-extracted, enzymatically bioconverted, or fermentation-derived.

The EFSA classification is particularly relevant for EU-market formulators. Reb A from leaf extraction falls under E 960a (steviol glycosides from stevia). Enzymatically bioconverted Reb M is classified separately as E 960c, with the EFSA FAF Panel confirming no safety concern and maintaining the established ADI of 4 mg/kg bw/day expressed as steviol equivalents.^[5] For a broader overview of how natural sweeteners navigate global regulatory frameworks, see our manufacturing guide.

Frequently Asked Questions

References