The ratio math (so you can improvise)
The tested recipes use protein at about 12.5%, allulose at 11–16%, and xanthan at about 0.4% of the milk weight. Those figures describe this recipe family; they are not universal ratios for every powder, liquid or frozen-dessert machine.
- Protein at 12.5% (50 g per 400 g milk). In this test set, lower amounts produced a thinner result and higher amounts emphasized powder flavor. Brands behave differently, so treat 50 g as the validated starting point.
- Allulose at 11–16% (45–65 g). It supplies sweetness, dissolved mass and freezing-point depression. More is not automatically better: higher amounts soften the base and also increase the per-serving gastrointestinal load.
- Xanthan at 0.4% (1.5 g). It increases viscosity and helps control ice crystals. Weigh it when possible; too much can produce a gummy texture.
- Fruit replaces part of the milk by weight. The tested fruit formulas reduce milk to stay below the MAX line and adjust allulose for the fruit's own sugar and solids. Use the pre-computed flavor recipes rather than assuming every fruit swaps 1:1.
The twelve flavors, one table
Per-serving macros, computed from USDA FoodData Central — click through for the full method:
| Flavor | kcal | Protein | Net carbs | Character |
|---|---|---|---|---|
| Vanilla Bean | 150 | 20.8 g | 1.9 g | The Classics |
| Chocolate Fudge | 169 | 21.8 g | 4.4 g | The Classics |
| Salted Caramel | 164 | 20.9 g | 2 g | The Classics |
| Strawberry | 163 | 21.2 g | 6.7 g | Fruit Scoops |
| Mint Chip | 199 | 21.7 g | 6.5 g | The Classics |
| Cold-Brew Coffee | 154 | 21.1 g | 3.6 g | The Classics |
| Almond Butter Swirl | 218 | 23.3 g | 2.7 g | The Classics |
| Banana Pudding | 191 | 21.3 g | 13.9 g | Fruit Scoops |
| Pumpkin Spice | 165 | 21.4 g | 5.4 g | Seasonal |
| Lemon Cheesecake | 192 | 22.1 g | 3.7 g | Dessert Case |
| Mango | 181 | 21.3 g | 13.1 g | Fruit Scoops |
| Piña Colada | 226 | 21.4 g | 10.5 g | Fruit Scoops |
The fix table (bookmark this)
| Pint looks like… | What actually happened | The fix |
|---|---|---|
| Crumbly or powdery after spin one | Common with low-fat bases on the first pass | Use Re-Spin. If it remains dry, add 1 Tbsp almond milk and Re-Spin again |
| Very dry snow | Base or freezer may be unusually cold | Let the pint rest briefly, then use Re-Spin; follow the instructions for your model |
| Icy and crunchy | Large crystals, slow freeze or altered ingredient ratios | Freeze a level pint for the full 24 hours and return to the tested formula |
| Chalky, tastes like powder | Undissolved protein or a strongly flavored powder | Blend a full 30 seconds, rest 5 minutes before freezing and check the powder label |
| Soft-serve soup | Under-frozen center or too much dissolved sweetener | Freeze the full 24 hours and use the tested allulose amount |
| Gummy or slimy | Too much xanthan | Use 1.5 g or a measured level ½ tsp |
| Firm after refreezing | Expected texture change after storage | Briefly temper and use Re-Spin if needed |
| Hump or hole in the frozen surface | Pint froze unevenly | Do not process an uneven solid block; follow the owner guide for your model |
Why the Creami is unusually good at this
Conventional churn machines build texture while the mixture freezes. The Creami instead processes a uniformly frozen pint after the freeze; Ninja describes the Creamerizer paddle as finely shaving and transforming the base. That makes it useful for lower-fat formulas, but the paddle does not erase the effects of ingredients or freezer conditions. The no-machinepan-scrape method is coarser and less repeatable, but it works without the appliance.