Gluten-free flour blends that actually work, and why single flours fail
Key takeaways
- Gluten did four jobs; blends replace it with a committee of flours, starches and a binder.
- A sound base ratio: 50% grain flour, 30% starch, 20% character flour — by weight.
- Rest GF batters 20–30 minutes and cool bakes fully before slicing.
- Xanthan for cakes, psyllium for breads — the binder choice follows the bake.
The first lesson of gluten-free baking is disappointing and liberating in equal measure: there is no single gluten-free flour that behaves like wheat flour, because wheat flour's magic is not its starch but its gluten — a protein network no rice, nut or root flour possesses. Every successful gluten-free blend is therefore an alliance: several flours and starches combining to imitate what one protein did alone.
This article explains what a blend must provide — structure, tenderness, moisture, flavour — which components provide each, a working all-purpose ratio to build from, and the handling differences that matter as much as the blend itself.
What the alliance must replace
Gluten did four jobs: it trapped gas (structure), stretched without tearing (elasticity), absorbed water in a particular way, and set with heat into crumb. A blend replaces them by committee. Rice and other grain flours give body and familiarity. Pure starches — cornflour, tapioca, potato starch — give lightness and tenderness, imitating the soft end of wheat. A binder such as xanthan gum or psyllium husk stands in for elasticity, holding gas and water in a web the flours cannot form alone. Protein-rich flours — sorghum, buckwheat, gram, nut flours — bring flavour and some scaffolding.
Remove any committee member and the bake shows it: all-rice blends are gritty and crumbly; all-starch blends are gummy; blends without a binder slump and slice like cake fudge. The proportions in commercial blends that work are not accidents; they are this committee, seated.
A working all-purpose ratio
A reliable starting blend for cakes, biscuits and general baking, by weight: 50% white rice flour (or a rice-and-sorghum base), 30% starch split between tapioca and cornflour or potato starch, and 20% of something characterful — sorghum, buckwheat, oat (certified gluten-free) or fine almond flour. Add xanthan gum at about ½ teaspoon per 150 g of blend for cakes and biscuits, and prefer psyllium (a teaspoon or two per loaf) for breads, where its gel holds a wetter, more bread-like dough.
Build by weight, always — gluten-free flours differ wildly in density, and cup measures of rice flour versus almond flour can be 40 g apart, which is why this site's flour guidance speaks in grams for blends. Make a kilogram at a time, whisk it thoroughly, and store it as your flour; consistency of blend is half the battle won before a recipe starts.
A practical note on cost, because gluten-free flours are not cheap and blends can feel extravagant. The economics improve sharply when you blend your own from bulk rice flour and starches — the two cheapest members of the committee — and reserve the expensive specialty flours (fine almond, certified oat) for the characterful fifth, where their flavour earns their price. Bins of failed single-flour experiments cost more than any blend; the ratio in this article, bought in bulk and mixed by the kilogram, lands close to the price of a good commercial blend with the advantage that you control every member of it. Keep the blend in a sealed tub, labelled with its ratio, and refills become a two-minute job rather than a research project.
Handling: where gluten-free diverges
Batters and doughs want more liquid than wheat intuition suggests, and they want rest: twenty to thirty minutes lets the flours hydrate and the gums organise, and many a ‘failed’ gluten-free cake was simply baked too soon after mixing. Cornflour-heavy blends can turn gummy if overmixed once wet; mix to combine, rest, and bake. Eggs carry more structural weight in this baking — an extra white or a well-whipped egg is a common and legitimate crutch in cakes.
Baking times run a little longer and lower for dense, moist blends, and doneness tests matter more than timers: gluten-free crumb sets late and can look done before it is. Cooling fully before slicing is not fussiness — the structure finishes setting as it cools, and a warm-sliced gluten-free loaf will gumminess itself into evidence against the whole enterprise.
Coeliac safety, briefly and seriously
For bakers avoiding gluten for medical reasons, blending your own means minding cross-contact: certified gluten-free oats and flours where the diagnosis demands it, clean scoops and jars, and no dipping a floury hand from the wheat bag into the blend. Rice flours and starches are naturally gluten-free but are milled in shared facilities often enough that the certified label matters for coeliac baking specifically.
None of this complexity should obscure the good news: the current generation of blends, built on the committee principle, produces cakes and breads that stand on their own merits. The goal is not a brave imitation of wheat — it is a good bake, judged as itself.
| Role | Flours / agents | Share of blend |
|---|---|---|
| Base & body | White rice, sorghum, millet flours | ~50% |
| Lightness | Tapioca, cornflour, potato starch | ~30% |
| Flavour & protein | Buckwheat, oat (certified GF), almond, gram | ~20% |
| Binder | Xanthan gum (cakes) / psyllium (breads) | ½ tsp per 150 g / 1–2 tsp per loaf |
The blend committee
The crumbly cake, recomposed
A first gluten-free sponge, made with pure rice flour, tastes fine and slices like sand. The diagnosis is committee-shaped: no starch for tenderness, no binder for cohesion. Rebuilt on the ratio above — rice base, tapioca and cornflour, a fifth almond flour, xanthan in the mix, batter rested twenty minutes — the same recipe slices cleanly, eats tender, and keeps a day longer. Nothing about the baker changed; the flour became a team.
Common mistakes
Using one flour and expecting wheat behaviour. A single rice or coconut flour cannot trap gas and stay tender alone. Blends exist because the jobs are plural.
Measuring blends by volume. GF flours’ densities diverge far more than wheat flours’. A cup-based blend is a different blend every time; weigh the components.
Skipping the rest before baking. Unhydrated GF flours bake gritty and dry. Twenty minutes of rest is an ingredient in all but name.
Questions, answered
Can I use a commercial blend in your ratio’s place?
Yes — a good commercial all-purpose GF blend already seats the committee. Add the binder only if the packet does not list one already included.
Why is my GF bread gummy inside?
Usually underbaked or sliced warm. Bake until the internal structure sets (a hollow knock, and a few extra minutes beyond), cool completely, then slice.
Is buckwheat flour gluten-free?
Despite the name, yes — buckwheat is unrelated to wheat. It brings protein and a deep flavour; use it as the characterful fifth of a blend.
The short version
Gluten-free baking stops being a consolation prize when the blend is designed rather than hoped for. Seat the committee, weigh it, rest it, and let the bakes be judged as bakes — which is the only judgement that ever mattered.
Put it into practice
The tool this article leans on: Flour Converter — Cups and grams for plain, bread, cake, wholemeal and other flours. Open it alongside “Gluten-free flour blends that actually work, and why single flours fail” and run the numbers for your own bake.
Open the Flour Converter →