Coconut flour: one third fiber, and the GI depends on how much you use

Baked goods made with a lot of coconut flour tested low on the glycemic index. Macaroons with 7.1 to 8.8 oz (200 to 250 g) of coconut flour per kg scored 45.7 and carrot cake 51.8, below the low-GI cut-off of 60, in a crossover test on 10 healthy and 10 diabetic adults eating single portions after an overnight fast. The abstract gives no separate figure for the diabetic group, and those foods also contained coconut fat. The same study found the opposite at low doses. Pan de sal with 50 g per kg scored 87.2 and a multigrain loaf with 100 g per kg scored 85.2, so a little coconut flour leaves ordinary bread a high-GI food. The flour itself is about one third fiber, 34.2 g per 3.5 oz (100 g) on average across 8 USDA samples, ranging from 30.9 to 39.5 g.

The numbers, per 3.5 oz (100 g)

How much is actually in it

• Coconut flour is about one third fiber: USDA samples average 34.2 g per 3.5 oz (100 g), ranging from 30.9 to 39.5 g. — Coconut flour sold in the US, 8 analyzed samples (n = 8) Evidence E source
  Total dietary fiber 34.24 g per 3.5 oz (100 g), range 30.9-39.5
• Coconut flour still carries coconut fat: USDA samples average 14.0 g saturated fat per 3.5 oz (100 g), varying from 9.2 to 18.1 g between samples. — Coconut flour sold in the US, 8 analyzed samples (n = 8) Evidence E source
  Saturated fat 13.999 g per 3.5 oz (100 g), range 9.24-18.14
• Coconut flour is rich in copper: USDA samples average 2.08 mg per 3.5 oz (100 g), more than twice the adult Daily Value of 0.9 mg. — Coconut flour sold in the US, 8 analyzed samples (n = 8) Evidence E source
  Copper 2.081 mg per 3.5 oz (100 g), range 1.65-2.24, 231% DV
• Coconut flour fiber is mostly mannan: mannose made up 75% of its fiber sugars in one lab analysis. — One commercial coconut flour, laboratory analysis (n = 3) Evidence D source
  Mannose 75.17%, glucose 14.42%, uronic acid 3.30% of fiber monosaccharides
• In one lab analysis coconut flour held 14.8% protein, 11.2% oil and 21.2% simple sugars. — One commercial coconut flour, laboratory analysis Evidence D source
  Protein 14.80%, oil 11.20%, total simple sugars 21.20%

What your body absorbs

• Lab data suggest coconut flour fiber survives digestion and is fermented by gut bacteria into as much butyrate as inulin. — Human fecal microbiota, in vitro fermentation after simulated upper gut digestion Evidence D source
  Butyrate similar to inulin, acetate:propionate:butyrate 70:12:18 vs 74:18:8 for inulin
• Lab data suggest coconut flour fiber ferments more slowly than inulin, with less butyrate in the first 6 hours. — Human fecal microbiota, in vitro fermentation Evidence D source
  Butyrate lower than inulin in the first 6 h (p < 0.05), similar by the end

What it does to your health

• In a double-blind crossover trial in 21 adults with raised cholesterol, corn flakes made with 25% coconut flakes, a by-product of coconut flour making, cut total cholesterol by 10.8% in 2 weeks, close to oat bran flakes at 8.4%. — 21 subjects with moderately raised serum cholesterol, four 2-week periods (n = 21) Evidence B source
  Total cholesterol -10.8% (SD 1.3) and LDL -9.2% (SD 5.4) with 25% coconut flakes, oat bran -8.4% and -8.8%

What people believe that the data does not support

• A little coconut flour does not make bread low-GI: pan de sal with 50 g per kg and a multigrain loaf with 100 g per kg had GI values of 87.2 and 85.2. — ten normal and ten diabetic subjects, single test meals (n = 20) Evidence B source
  GI 87.2 (SEM 5.5) pan de sal, 85.2 (SEM 6.8) multigrain loaf

Who this works differently for

• In a randomized crossover test in 10 healthy and 10 diabetic adults, macaroons and carrot cake made with 200-250 g coconut flour per kg had low glycemic index values of 45.7 and 51.8. — ten normal and ten diabetic subjects, single test meals after an overnight fast (n = 20) Evidence B source
  GI 45.7 (SEM 3.0) macaroons, 51.8 (SEM 3.3) carrot cake, below the low-GI cut-off of 60

What is still unknown

• In the same 21-person trial triglycerides also fell 14.5% on plain corn flakes, so the 21.8% fall with 25% coconut flakes cannot be credited to coconut fiber alone. — 21 subjects with moderately raised serum cholesterol, triglycerides analyzed only in the 60% with levels above 170 mg/dL (n = 21) Evidence B source
  Triglycerides -14.5% corn flakes control, -21.8% 25% coconut flakes

The bottom line

It is not a fat-free flour. The 8 USDA samples averaged 14.0 g saturated fat per 3.5 oz (100 g), and the spread from 9.2 to 18.1 g likely reflects how far each flour was defatted. Copper is high at 2.08 mg per 3.5 oz (100 g), about 231 percent of the Daily Value, though a recipe rarely uses 100 g per person and the US survey has no standard portion for this flour. One laboratory sample held 14.8 percent protein, 11.2 percent oil and 21.2 percent simple sugars, and that sugar figure should not be read as typical. The cholesterol evidence comes from a related product. In a double-blind crossover trial, 21 adults with moderately raised cholesterol ate corn flakes made with 25 percent coconut flakes, a by-product of making the flour. Total cholesterol fell 10.8 percent and LDL 9.2 percent in 2 weeks, close to oat bran flakes at 8.4 and 8.8 percent. We found no larger or longer repeat. Triglycerides are less clear. They fell 21.8 percent on the coconut flakes, but plain corn flakes also cut them by 14.5 percent, and only the 60 percent of people with levels above 170 mg/dL were analyzed. The drop cannot be credited to coconut fiber alone. Lab data suggest the fiber, mostly mannan with mannose making up 75 percent of its sugars in one sample, is fermented by human gut bacteria in a dish into as much butyrate as inulin by the end, only more slowly in the first 6 hours. We found no trial testing this in people. We found no study of coconut flour in allergy, pregnancy, children or with medicines, so this page has no safety section.

Sources

1. cfl-daily-r4-05 · reference dataset
   FDC 2515382 Flour, coconut: Fiber, total dietary 34.24 g per 100 g (Foundation samples 30.9-39.5, n=8)
   USDA FoodData Central Foundation Foods, FDC 2515382 (Flour, coconut) · checked 2026-09-29
2. cfl-daily-r4-06 · reference dataset
   FDC 2515382 Flour, coconut: Fatty acids, total saturated 13.999 g per 100 g (Foundation samples 9.24-18.14, n=8)
   USDA FoodData Central Foundation Foods, FDC 2515382 (Flour, coconut) · checked 2026-09-29
3. cfl-daily-r4-07 · reference dataset
   FDC 2515382 Flour, coconut: Copper, Cu 2.081 mg per 100 g (Foundation samples 1.65-2.24, n=8)
   USDA FoodData Central Foundation Foods, FDC 2515382 (Flour, coconut) · checked 2026-09-29
4. cfl-daily-r4-10 · laboratory analysis
   The monosaccharide composition analyses revealed that coconut flour DFs are dominated by mannose residues (75.17%), followed by glucose (14.42%), uronic acid (3.30%), galactose (3.11%), xylose (2.51%), and arabinose (1.08%) (Figure 1), indicating that coconut flour DFs are composed mainly of mannans.
   Food Science & Nutrition, 2025 · checked 2026-09-29
5. cfl-daily-r4-11 · laboratory analysis
   Coconut flour was found to contain significantly (p < 0.05) higher protein (14.80%) and oil (11.20%) contents, compared to carob flour (3.73% and 0.36%, respectively), whereas carob flour was found to possess significantly (p < 0.05) higher total simple sugar content (31.62%), compared to coconut flour (21.20%) (Table 1).
   Food Science & Nutrition, 2025 · checked 2026-09-29
6. cfl-daily-r4-08 · in vitro study
   Our results revealed that fermentation of coconut flour DFs resulted in a generation of similar amount of butyrate that was observed in inulin, indicating that coconut flour DFs is as butyrogenic as inulin. In addition, the acetate:propionate:butyrate ratio was 74:18:8 at the end of inulin fermentation, whereas that was 70:12:18 at the end of coconut flour DFs fermentation, further confirming the high butyrate
   Food Science & Nutrition, 2025 · checked 2026-09-29
7. cfl-daily-r4-09 · in vitro study
   Interestingly, at early stage of the fermentation (in the first 6 h of fermentation), the butyrate level observed in inulin sample was significantly (p < 0.05) higher than that was observed in coconut flour DFs sample, suggesting that butyrate generation by colonic microbiota as results of coconut flour DFs fermentation occurs at a slower rate, compared to that of inulin.
   Food Science & Nutrition, 2025 · checked 2026-09-29
8. cfl-daily-r4-03 · randomized crossover trial
   The study was conducted in a double-blind randomized crossover design on a 14-week period, consisting of four 2-week experimental periods, with each experimental period separated by a 2-week washout period. ... Results showed a significant percent reduction in serum total and low-density lipoprotein (LDL) cholesterol (in mg/dL) for all test foods, except for corn flakes, as follows: oat bran flakes, 8.4 +/- 1.4 and 8.8 +/- 6.0, respectively; 15% coconut flakes, 6.9 +/- 1.1 and 11.0 +/- 4.0, respectively; and 25% coconut flakes, 10.8 +/- 1.3 and 9.2 +/- 5.4, respectively.
   J Med Food, 2004 · checked 2026-09-29
9. cfl-daily-r4-02 · randomized crossover trial
   Among the test foods, pan de sal (GI 87.2 (sem 5.5)) and multigrain loaf (GI 85.2 (sem 6.8)) gave significantly higher GI with 50 and 100 g coconut flour/kg respectively (P<0.05).
   Br J Nutr, 2003 · checked 2026-09-29
10. cfl-daily-r4-01 · randomized crossover trial
    The significantly low-GI (<60) foods investigated were: macaroons (GI 45.7 (sem 3.0)) and carrot cake (GI 51.8 (sem 3.3)), with 200-250 g coconut flour/kg (P<0.05).
    Br J Nutr, 2003 · checked 2026-09-29
11. cfl-daily-r4-04 · randomized crossover trial
    Serum triglycerides were significantly reduced for all test foods: corn flakes, 14.5 +/- 6.3%; oat bran flakes, 22.7 +/- 2.9%; 15% coconut flakes, 19.3 +/- 5.7%; and 25% coconut flakes, 21.8 +/- 6.0%. Only 60% of the subjects were considered for serum triglycerides reduction (serum triglycerides >170 mg/dL).
    J Med Food, 2004 · checked 2026-09-29