What is Glycemic Index (GI)?
The Glycemic Index ranks carbohydrate-containing foods on a scale of 0 to 100, based on how quickly they raise blood glucose compared to pure glucose, which scores 100. The faster a food raises blood glucose, the higher its GI.
A food is tested by having volunteers eat a fixed amount (containing 50 g of available carbohydrate), then measuring blood glucose over two hours. The area under the glucose curve is compared to the same person's response to 50 g of pure glucose. The result — averaged across several people — is the GI value.
≤ 55
Slow, steady rise
56 – 69
Moderate rise
≥ 70
Fast, sharp rise
What affects a food's GI?
GI is not fixed — it shifts depending on how a food is prepared and what you eat it with:
- Cooking and processing — boiled potatoes have a lower GI than mashed or baked ones. Al dente pasta has a lower GI than overcooked pasta.
- Ripeness — a ripe banana has a higher GI than an unripe one, as starches convert to sugars as the fruit matures.
- Fibre content — soluble fibre (oats, lentils, apples) slows glucose absorption and lowers effective GI.
- Protein and fat in the same meal — eating carbohydrates alongside protein and healthy fats slows gastric emptying and reduces the GI of the whole meal.
- Acidity — vinegar or lemon juice added to a meal can measurably lower its GI response.
What is Glycemic Load (GL)?
Glycemic Load solves the most important limitation of GI: it only matters how fast glucose enters the blood if you are actually eating enough carbohydrate for it to make a difference. GI is measured using a fixed 50 g carbohydrate dose — but real serving sizes vary enormously.
GL adjusts GI for the actual amount of carbohydrate in a typical serving:
≤ 10
Modest glucose impact
11 – 19
Moderate impact
≥ 20
Significant glucose load
GL is the more useful number for real-world food choices. It tells you the actual glucose burden a serving of food places on your body — not just how fast it would raise glucose in a lab experiment.
Why GI and GL Can Tell Different Stories
This is where it gets interesting — and where people get confused. A food can have a low GI but a high GL, or a high GI but a low GL. Two examples make this clear:
Watermelon — High GI, Low GL
Watermelon has a GI of around 72 — firmly in the "high" category, which alarms people. But a standard 200 g serving of watermelon contains only about 6 g of available carbohydrate (it is mostly water). Plugging this into the formula: 72 × 6 ÷ 100 = GL of 4 — very low. The glucose response from eating a slice of watermelon is genuinely modest. GI alone was misleading.
Red Lentil Dal with Brown Rice — Low GI, High GL
Red lentils have a GI of around 29 and brown rice around 55 — both solidly low-to-medium GI. But a full serving of dal with rice can contain 80–100 g of carbohydrate. At a weighted GI of around 49: 49 × 90 ÷ 100 = GL of 44 — very high. The glucose rise will be slow and sustained rather than a spike, but there will be a significant total glucose load. For someone managing insulin doses, this matters just as much as a high-GI meal.
Key insight: Low GI does not mean low GL. When eating carb-dense foods like rice, legumes, oats or bread — even wholefood versions — GL tells you more than GI. A slow glucose rise is beneficial, but a large sustained rise still requires an insulin response.
GI, GL and Blood Sugar Management
For anyone managing blood glucose — whether through diet alone, oral medication or insulin — the practical implications of GL are significant:
- Low GL meals produce a gentler, more gradual glucose rise, which reduces the insulin response required and helps avoid energy crashes or hypoglycaemia.
- High GL meals produce a larger total glucose load — even if the rise is slow (low GI). If you are carbohydrate-counting for insulin dosing, the total carbohydrate figure is what matters, not just the GI.
- Combining GI and GL gives you the full picture: a low-GI, low-GL meal is ideal for stable blood glucose. A low-GI, high-GL meal still needs careful portion management.
Practical ways to reduce GL
- Reduce the portion of the high-carb component (rice, pasta, potato, bread) and increase non-starchy vegetables, which add volume and fibre with negligible GL.
- Eat carbohydrates alongside protein and healthy fat — this slows digestion and lowers the meal's overall glycemic response.
- Choose less-processed wholefood versions: brown rice over white, whole oats over instant, lentils over white bread.
- Add acidity — a squeeze of lemon or a small amount of vinegar in a dressing measurably lowers GI response.
- Do not overcook starchy foods — al dente pasta and firm potatoes have lower GI than soft, overcooked versions.
GI & GL of Common Wholefoods
Values below are from the International Tables of Glycemic Index and Glycemic Load Values (Atkinson et al., 2021). GL is calculated for a typical serving size.
| Food | GI | Typical serving | Carbs (g) | GL | GL rating |
|---|---|---|---|---|---|
| Rolled oats (porridge) | 55 | 40 g dry | 27 | 15 | Medium |
| Brown rice (boiled) | 55 | 150 g cooked | 33 | 18 | Medium |
| Basmati rice (white) | 60 | 150 g cooked | 45 | 27 | High |
| Red lentils (boiled) | 29 | 150 g cooked | 15 | 4 | Low |
| Chickpeas (boiled) | 28 | 150 g cooked | 27 | 8 | Low |
| Banana (ripe) | 53 | 120 g (1 medium) | 23 | 12 | Medium |
| Apple | 36 | 120 g (1 medium) | 15 | 5 | Low |
| Sweet potato (boiled) | 44 | 150 g | 20 | 9 | Low |
| Watermelon | 72 | 200 g (1 slice) | 6 | 4 | Low |
| Quinoa (boiled) | 53 | 150 g cooked | 30 | 16 | Medium |
| Wholegrain bread | 51 | 30 g (1 slice) | 13 | 7 | Low |
| White bread | 75 | 30 g (1 slice) | 14 | 11 | Medium |
| Carrots (boiled) | 39 | 80 g | 4 | 2 | Low |
Important Limitations
GI and GL values are population averages. Individual glucose responses vary significantly based on gut microbiome composition, physical fitness, sleep, stress levels and medication. A food that produces a gentle glucose rise in one person may cause a sharp spike in another.
The values in published tables are measured from a single food in isolation. In practice, you eat mixed meals — and protein, fat and fibre from other components of the meal substantially alter the glycemic response of every ingredient. A recipe's GI/GL figure is a useful guide, not a precise prediction.
GI and GL say nothing about a food's broader nutritional quality. White sugar has a medium GI (65). Carrots have a high-ish GI (39–49 depending on cooking) but are a nutrient-dense vegetable. Use GI/GL alongside — not instead of — overall nutritional context.
Why Some Foods Don't Have a GI or GL Value
If you browse the Nutriofia food library, you will notice that GI and GL data is present for roughly 277 foods — about one in five. This is not a gap in our database; it reflects genuine constraints in how GI is measured and which foods science has tested.
GI testing requires human volunteers in controlled conditions
Measuring a food's GI is not a simple lab procedure. It requires real people to eat a fixed portion of the food after an overnight fast, then have their blood glucose measured at regular intervals over two hours. Each food needs multiple participants and multiple test sessions to produce a reliable average. This makes GI research expensive and slow — only commercially significant or clinically interesting foods tend to get tested.
Many foods cannot have a GI value
GI only applies to foods that contain meaningful amounts of carbohydrate. If a food contains fewer than around 5–10 g of available carbohydrate per typical serving, there is not enough glucose entering the bloodstream to produce a measurable response. This means the following categories simply cannot have a GI or GL:
- Oils and fats — olive oil, coconut oil, butter: zero carbohydrate, no glucose response.
- Most non-starchy vegetables — broccoli, spinach, courgette, kale, peppers: carbohydrate content too low to test meaningfully.
- Meat, fish and eggs — protein and fat only, no carbohydrate.
- Most herbs and spices — used in tiny quantities with negligible carb content.
- Cheese — very low carbohydrate; protein and fat only.
For these foods, the absence of a GI value is a feature, not a flaw — it tells you the food does not raise blood glucose at all in normal portions.
Published data does not cover every food
Even among carbohydrate-containing foods, GI data exists only for those that have been formally tested and published in peer-reviewed journals. Many traditional, regional or less commercially prominent foods simply have never been studied. The most comprehensive published source — the International Tables used by Nutriofia — contains around 4,000 entries covering common staples worldwide, with the highest-quality ISO-standard data available for about 2,100 of those.
If a food you eat regularly doesn't have a GI or GL on Nutriofia, the most practical approach is to look at its total carbohydrate content per serving. Foods with fewer than 10 g of available carbs per serving will have a low GL regardless of their GI — and foods with very low carb content are unlikely to affect your blood glucose significantly.
On Nutriofia, GI and GL values are sourced from the open-access International Tables of Glycemic Index and Glycemic Load Values 2021 (Atkinson et al., Am J Clin Nutr), which compiles ISO-standard measurements from research laboratories worldwide. Recipe GI is calculated as a carbohydrate-weighted average across all ingredients with GI data.
Source: Atkinson FS, Brand-Miller JC, Foster-Powell K, Buyken AE, Goletzke J. International tables of glycemic index and glycemic load values 2021: a systematic review. Am J Clin Nutr. 2021;114(5):1625-1632. doi:10.1093/ajcn/nqab233
GI and GL values are population averages from controlled studies. Individual responses vary. This article is for educational purposes and does not constitute medical advice. If you manage a health condition including diabetes or insulin resistance, discuss dietary changes with your healthcare team.