Skip to content
metaboliq

How many carbs a day on a keto diet?

  • The short answer first, then the detail.
  • 8 sources, each linked in full.
  • A registered doctor reads your own results.
Book my health assessment

Takes 2 minutes to book. A doctor makes every decision with you.

A young woman with her hair in a claw clip laughing at a sunny dining table over a colourful plate of bajra roti, paneer, dal and salad

Keep net carbs between 20 and 50 grams a day. Net carbs are the total carbohydrate on the label minus the fibre. That is the range the American Diabetes Association gives for a very low-carb diet meant to bring on ketosis, and most beginners start at the bottom of it. To see how tight that is, one katori of cooked rice holds about 40 grams. One serving at lunch and the day is nearly spent.

If you are doing keto to lose weight, the exact number matters less than the forums make out. In a big year-long American trial, people eating about 30% of their calories as carbs lost as much as a low-fat group. So choose a limit you can still keep when there is a wedding, a birthday and a work dinner in the same week.

What is the carb limit on a keto diet?

The figure you will see everywhere is 20–50 g of net carbs a day, and it has a solid basis. The American Diabetes Association's 2019 consensus report, led by Evert, describes very low-carb eating as aiming for 20–50 g of nonfiber carbohydrate to bring on ketosis. Fat supplies more than half the calories.

Fifty grams is also the cut-off researchers use. When Bueno and colleagues gathered the year-long trials of keto against low-fat diets, a diet counted as ketogenic only at 50 g a day or less.

Above that, you are eating low carb rather than keto. That can still be a very good diet. A woman on 1,500 calories who eats 80–100 g of carbs, say two rotis, a katori of dal and a fruit, is well inside the ADA's low-carb range. She may lose weight steadily, but a ketone meter will mostly read close to zero.

Why does the line sit near 50 g? The liver holds only about a day's supply of glucose. Eat less than that for two or three days and the store runs down, so the liver turns to fat and makes ketones, which the brain and muscles then burn. Westman's 2007 review describes the switch and the fall in appetite that often follows.

Sources: [1], [2], [3]

Should you count net carbs or total carbs?

Most keto plans count net carbs: total carbohydrate minus dietary fibre. Fibre is not absorbed as sugar, so it barely touches blood glucose or ketone levels. The ADA report's phrase "nonfiber carbohydrate" is the same idea in formal language.

For Indian food the difference is real. A 100 g bowl of cooked palak or methi carries only a gram or two of net carbs once the fibre is taken off. A bowl of rajma or chole keeps a good share of its carbs even after that. Counting net carbs lets you eat a large plate of leafy sabzi without blowing the limit, and that keeps meals filling.

Two cautions. First, packaged "keto" snacks sold in India often subtract sugar alcohols as well as fibre to print a low net figure; read the total carbohydrate line and judge for yourself. Second, if a doctor is using ketone readings to guide you, ask which way they want you to count, because some clinics work from total carbs to keep things simple.

If you are new to this and want one rule, use net carbs, keep the first two weeks at 20–30 g, and write down everything, including the sugar in chai.

Sources: [1]

What does 20–50 g of carbs look like on an Indian plate?

The numbers below are rough net-carb figures for home portions, using a 150 ml katori. Recipes and brands vary, so check a label or a food table when you can.

One medium phulka made with about 30 g of atta has roughly 18 g. A katori of cooked rice has about 40 g, and a katori of cooked toor dal about 12–15 g. A medium potato or a medium banana each come to about 25 g. A glass of milk has about 10 g, and a cup of chai with two spoons of sugar adds another 10 g.

At the low end sit paneer, at a few grams per 100 g, and eggs, at under a gram each. Chicken, fish and mutton are close to zero. Plain curd is about 4–5 g per 100 g, and leafy or watery vegetables barely register. A small handful of almonds or walnuts adds 2–3 g.

A day at about 25 g might run like this. Breakfast is a two-egg bhurji with half an onion and a tomato, cooked in ghee, at about 5 g. Lunch is a katori of palak paneer, a bowl of cucumber and a small katori of curd, at about 10 g. Dinner is 150 g of tandoori chicken or grilled fish with a katori of cauliflower sabzi, at about 8 g. Chai goes without sugar, with a splash of milk adding a gram or two.

A vegetarian day works the same way, with paneer, tofu, curd and nuts carrying the protein. What has to go, at least at first, is rice, roti, large portions of dal, potatoes, sweet fruit and anything with sugar in it.

Does everyone need the same carb limit?

No. The number that keeps one person in ketosis may be too high for another, and insulin resistance is a large part of why.

In the Virta Health study reported by Hallberg and colleagues, adults with type 2 diabetes were coached to reach nutritional ketosis. The authors note that, with the insulin resistance typical of type 2 diabetes, people usually needed total carbohydrate below 30 g a day to get there. Coaches then adjusted each person's limit by their daily ketone readings.

Activity matters too. Someone who walks 12,000 steps a day or trains hard burns through stored glucose faster and may stay in ketosis at 40–50 g. A person at a desk with a large waist may need to stay near 20 g.

Indian adults tend to start from a high-carb baseline and a high rate of insulin resistance. Anjana and colleagues modelled diets from the national ICMR-INDIAB survey. The patterns most likely to bring blood sugar back to normal still took about half their calories from carbohydrate, with more protein than people usually eat. That is far above keto, which is a useful reminder that ketosis is one tool, not the only route to better sugar control.

Sources: [4], [5]

How do you know if you are in ketosis?

The surest home check is a blood ketone meter, the finger-prick kind that looks like a glucose meter. It reads beta-hydroxybutyrate. In the Hallberg study, people aimed for 0.5–3.0 mmol/L, the band usually called nutritional ketosis. Test first thing in the morning, before tea, on three or four separate days. A single reading after a hard workout or a heavy dinner can mislead you either way.

Urine strips are cheaper and every chemist has them. They work well in the first two or three weeks, when your body is spilling spare ketones into urine. After a month they often turn pale even while blood ketones stay up, because your muscles have learned to use them. A faded strip in week six is not, by itself, a sign that you slipped.

Plenty of people use no meter at all. They notice that they are not hungry by eleven, that their mouth feels dry or tastes faintly of metal, and that the scale has dropped a kilo or two in the first week.

Most of that first drop is water. Stored glycogen holds a lot of water, and keto empties those stores within days. Fat loss starts to show around week three, which is when the weekly average on the scale becomes worth watching.

Sources: [4]

Does a stricter carb limit burn more fat?

It feels as though it should. Fewer carbs, more fat burned, faster results. When researchers have measured it carefully, though, the extra burn turns out to be tiny.

Hall's group at the US National Institutes of Health kept 17 overweight men on a metabolic ward for two months. For two days each week they lived in a sealed room that measured the calories they used. After a month of high-carb meals, their carbs were swapped for fat at exactly the same calories. In that room, keto raised their burn by 57–89 calories a day; a tracer-drink method over the last two weeks put the rise at about 151. Both are small numbers. Fat loss slowed, if anything.

Over a year, the gap between strict and relaxed carb limits is small. Bueno's meta-analysis put people on very low-carb diets about 0.9 kg lighter than low-fat groups after a year or more. In the DIETFITS trial, 609 adults on healthy low-carb or healthy low-fat diets ended the year with similar losses, though the low-carb group had drifted up to about 30% of calories from carbs.

So a strict limit helps only if it makes you eat less, and for many people it does. Hunger goes quiet and the evening snack stops. For others it backfires: in a later inpatient study by Hall's team, volunteers ate about 689 more calories a day on a keto menu than on a low-fat plant menu. Give it three weeks, and if you are hungrier and the weekly average has not moved, a looser limit may suit you better.

Sources: [6], [2], [7], [8]

What happens after the first few months on keto?

Very few people stay at 20 g for good, and nobody needs to. A common way out is two or three strict months, then adding 5–10 g of carbs a week. A small katori of dal one week, a fruit the next. Each Sunday, check the weekly scale average, the tape around your waist and how hungry you felt. Stop adding when those start to slip.

Where you level off depends on how your body deals with carbs. A man whose fasting insulin and HbA1c started high may settle around 50–80 g. Someone with normal results may get back above 100 g with an occasional roti. The ADA report treats all of these as healthy low-carb patterns.

Muscle is what most people forget to protect. In Hall's ward study the men lost more lean tissue in the keto month, and a bathroom scale cannot tell you which kind of kilo went. This is why, at Metaboliq, a doctor reads a home blood test for sugar, insulin and lipids before setting anyone's carb level. The plan aims to lose up to 20% of body weight while keeping muscle, so the carb number comes with protein at every meal and two strength sessions a week.

Repeat your lipid profile at about three months. In Bueno's analysis, LDL cholesterol crept up slightly on very low-carb diets while HDL rose more, and your own numbers could move either way.

Sources: [1], [6], [2]

Who should not go below 50 g of carbs without a doctor?

If you are pregnant, have chronic kidney disease or have ever had an eating disorder, see a doctor before you start. The ADA report says the research is still too thin to recommend very low-carb eating in any of these situations.

If you take insulin or anything else that lowers sugar, the first week is the risky part. Readings can drop within two or three days of cutting rice and roti. Your doctor will usually scale back sugar-lowering treatment before you start, and ask you to check your sugar more often. One healthy volunteer in Hall's 2021 study had to leave the keto arm with low sugar, so this is not only a worry for diabetics.

You may hear that keto causes ketoacidosis. That is a different thing: an emergency where ketones and blood sugar both run very high, seen mainly in type 1 diabetes and with some diabetes treatments. Food ketosis in a healthy adult stays far below those levels.

People on blood pressure treatment often feel light-headed in the first fortnight, getting up from the floor or out of bed, because the body sheds salt and water. Sit on the edge of the bed for a moment before you stand, and tell your doctor if it keeps happening. Everyone else can usually start safely. Drink a couple of extra glasses of water, keep salting your sabzi normally and fill half the plate with greens.

Sources: [1], [8]

Every source, in full

Linked to the publisher or to the abstract, so you can read them yourself.

  1. [1] Evert AB, Dennison M, Gardner CD, et al. Nutrition Therapy for Adults With Diabetes or Prediabetes: A Consensus Report

    Diabetes Care, 2019. Describes very-low-carbohydrate eating as often aiming for 20–50 g of nonfiber carbohydrate a day to induce nutritional ketosis, defines it as under 26% of calories, and calls for more research before advising it in pregnancy, kidney disease or disordered eating. Accessed 28 October 2026.

    Read the source
  2. [2] Bueno NB, de Melo IS, de Oliveira SL, da Rocha Ataide T. Very-low-carbohydrate ketogenic diet v. low-fat diet for long-term weight loss: a meta-analysis of randomised controlled trials

    British Journal of Nutrition, 2013. Defined a very-low-carbohydrate ketogenic diet as no more than 50 g of carbohydrate a day; at 12 months or more it gave about 0.9 kg more loss than low-fat diets, with higher HDL and slightly higher LDL. Accessed 28 October 2026.

    Read the source
  3. [3] Westman EC, Feinman RD, Mavropoulos JC, et al. Low-carbohydrate nutrition and metabolism

    American Journal of Clinical Nutrition, 2007. Under carbohydrate restriction, fuel use shifts from glucose and fatty acids to fatty acids and ketones, and unrestricted low-carbohydrate diets tend to reduce appetite. Accessed 28 October 2026.

    Read the source
  4. [4] Hallberg SJ, McKenzie AL, Williams PT, et al. Effectiveness and Safety of a Novel Care Model for the Management of Type 2 Diabetes at 1 Year: An Open-Label, Non-Randomized, Controlled Study

    Diabetes Therapy, 2018. Participants aimed for a blood beta-hydroxybutyrate level of 0.5–3.0 mmol/L; with type 2 diabetes, total carbohydrate usually had to be under 30 g a day to reach nutritional ketosis. Accessed 28 October 2026.

    Read the source
  5. [5] Anjana RM, Srinivasan S, Sudha V, et al. Macronutrient Recommendations for Remission and Prevention of Diabetes in Asian Indians Based on a Data-Driven Optimization Model: The ICMR-INDIAB National Study

    Diabetes Care, 2022. In 18,090 Indian adults, diet patterns modelled for diabetes remission took about 49–54% of energy from carbohydrate and 19–20% from protein, recommending less carbohydrate and more protein than usual. Accessed 28 October 2026.

    Read the source
  6. [6] Hall KD, Chen KY, Guo J, et al. Energy expenditure and body composition changes after an isocaloric ketogenic diet in overweight and obese men

    American Journal of Clinical Nutrition, 2016. Switching 17 men to a ketogenic diet at the same calories raised energy expenditure only slightly; body fat loss slowed and fat-free mass loss increased during the ketogenic phase. Accessed 28 October 2026.

    Read the source
  7. [7] Gardner CD, Trepanowski JF, Del Gobbo LC, et al. Effect of Low-Fat vs Low-Carbohydrate Diet on 12-Month Weight Loss in Overweight Adults and the Association With Genotype Pattern or Insulin Secretion: The DIETFITS Randomized Clinical Trial

    JAMA, 2018. In 609 adults, a healthy low-carbohydrate diet averaging 30% of energy from carbohydrate and a healthy low-fat diet produced similar weight change at 12 months. Accessed 28 October 2026.

    Read the source
  8. [8] Hall KD, Guo J, Courville AB, et al. Effect of a plant-based, low-fat diet versus an animal-based, ketogenic diet on ad libitum energy intake

    Nature Medicine, 2021. In 20 inpatients eating freely, a low-fat plant-based menu led to about 689 fewer calories a day than an animal-based ketogenic menu; one volunteer withdrew from the ketogenic arm with low blood sugar. Accessed 28 October 2026.

    Read the source

Your plan starts with a health assessment.

A carb limit is a guess until you know how your body handles glucose. Fasting insulin, HbA1c and lipids let a doctor set a carb level that takes weight off while you keep your muscle.

Book my health assessment

Then the home test, ₹999: blood test, 45+ metabolic markers, collected at your home.

Takes 2 minutes to book. A doctor makes every decision with you.

Book my health assessment
Book my health assessment

Takes 2 minutes to book. A doctor makes every decision with you.