Cook time calculator

Why "15–20 minutes" charts don't work

Search a cook time for almost any food and you'll get a chart claiming one number works for every air fryer. Owners know better. Under the most-shared cook-time cheat sheets, the top comment is reliably some version of "every air fryer is completely different — you need to learn your own machine," and the chart those same readers rate highest is the one that names its wattage and basket size up front.

The failures lean one way, too. Burnt and overdone food comes up far more often in owner complaints than undercooked food does, which tells you what a generic chart is really doing: not erring towards a slow, safe number, but printing whatever time happened to work in the machine its author owns.

So we stopped looking for a better generic number — there isn't one. For each appliance and food pair, this calculator shows the value that appliance's own manual states, a value we tested directly, or an honest "we don't have this yet". Never a guess dressed up as data.

Our method: manuals, cited page by page, never averaged

Every measurement here traces back to a manual page, a test we ran, or a moderated reader submission — the badge on each result says which, and how we test sets out the full method.

The rule that does the most work is the one about disagreement. We never average two appliances together to produce a smoother, more presentable number. When two brands disagree — and they frequently do — both values stay visible above, each carrying its own citation, rather than being blended into a figure that appears in neither manual and belongs to neither machine.

The four variables a generic chart can't see

A flat "15–20 minutes" ignores four things. An appliance-specific chart accounts for all four, because it was written for one machine and nothing else.

Wattage. More power should mean faster cooking, and sometimes it does — but not reliably, and not for every food. Tower sells essentially the same air fryer design at 1000 W, 1400 W, and 2600 W, with an official chart for each. The jump from 1000 W to 1400 W changed nothing at all across six shared foods in their own charts; the jump to 2600 W cut chicken breast and fries times by roughly a quarter, but left frozen nuggets untouched. See our full breakdown of Tower's own wattage numbers for the complete comparison — wattage alone predicts almost nothing without knowing which food you're asking about.

Basket format. Single basket, dual zone, and oven-style units aren't small variations on the same shape — they move hot air differently around the food, and they hold very different amounts of it. Comparing every case where a manufacturer publishes a time for the same food at the same temperature in both shapes, oven-style charts are never quicker than basket charts and are usually only a minute or two slower, while carrying substantially bigger batches — the full comparison is on basket versus oven-style, and what a second drawer changes answers the dual-zone half. That's why the specs picker above asks for basket format before it asks anything else: it's the axis people actually know about their own appliance.

Thickness and weight. A manual's stated time is only valid for the weight range it was tested against. A single chicken breast and four stacked chicken breasts don't cook in the same time at the identical temperature, and the assumptions line under any result above flags this directly whenever the underlying food is sensitive to it.

Preheat. Some manuals build preheating into their stated time; others explicitly skip it. A chart that assumes the basket is already hot reads fast next to one that includes warm-up — same appliance, same displayed temperature, different starting condition. So the calculator states, for the exact appliance you picked, whether preheat is part of the assumption instead of leaving it hidden.

These four don't interact simply enough to fold into a single adjustment factor. One wattage-based scalar would overstate the benefit of Tower's 2600 W model for frozen nuggets and understate it for chicken breast. That's why our schema keeps a wattage-driven time factor as an optional, per-appliance field rather than a formula applied to every food on the strength of one product line's numbers.

Four real manuals, one food, one near-identical temperature

The table below isn't hypothetical. It's four official manuals, each giving a time forchicken breast at 180–200°C, read directly from the same measurement records the calculator above uses.

Sources (official manuals/product pages, one per appliance): Gourmia 8-Qt Digital Air Fryer, Gourmia 4-Qt Digital Air Fryer, Chefman TurboFry Touch Dual Air Fryer, Midea 11 qt. 10-in-1 Dual-Zone Air Fryer Oven. Full citation detail on each appliance page.

Gourmia settles it without leaving the brand. Its 8-quart GAF858 and its 4-quart GAF414 both carry a recipe book with the same cooking chart, the same poultry row, the same two 8 oz breasts — and the charts say 450°F for one machine and 400°F for the other. Both then ask for 15–20 minutes. One manufacturer, one chart layout, a 50°F difference.

Wattage doesn't order them either. Three of these four draw an identical 1700 W and still span 90°F: the Gourmia GAF858 at 450°F, the Chefman dual zone at 375°F, the Midea at 360°F. The coolest of them wants the longest time — 21–22 minutes against 15–20 — while the 1300 W GAF414 is set hotter than either 1700 W dual-zone machine. What actually separates them isn't on the spec sheet: basket geometry and how each manufacturer calibrated its own chart are both plausible, and no manual says which.

None of the four is "wrong". Each number is what that manufacturer states for that machine, sourced to the page cited above. Averaging them into "18 minutes for chicken breast" would produce a figure that's wrong for all four at once.

Using this calculator without guessing

Start with brand and basket format. Most people know those two things about their air fryer even when the model code means nothing to them, which is why that's the default path above rather than a dropdown of model numbers. If your model has a direct measurement on file, you'll get its sourced number and a link to the manual page it came from. If it doesn't, you'll get a clearly labelled generic reference — a labelled estimate you can trust beats a precise-looking one you can't.

Whatever number comes back, treat the clock as the starting point and the internal temperature as the answer, for meat, poultry and fish. The USDA thresholds below don't move for any appliance. Check a few minutes early rather than trusting a timer, ours included, down to the last second.