Scope. This page provides a controlled reader-run protocol and does not claim site-run measurements. For cooking, cooling, storage and reheating, follow the current official guidance linked on the page.
What this page is, and what it is not
This is a signposted summary of other people's authority. It is not our guidance.
Every number below belongs to the named food-safety authority beside it and carries a retrieval date so that you can check whether it has since changed. We are not the source. We are an index with citations attached, and the value of an index is that it sends you to the original rather than replacing it.
That distinction matters more here than anywhere else on the site. Elsewhere on the site, a texture question can be answered with a probe and a log. Storage duration is not a thing a kitchen can measure. Establishing how long a cooked food remains safe under refrigeration requires microbiological work, controlled conditions, and a population of samples, and no amount of careful cooking substitutes for that. So on this subject we have no findings of our own and we are not going to pretend otherwise.
The reason this page exists is that the alternative is worse. Storage advice online is produced with total confidence and almost no attribution. Numbers circulate, get rounded, get copied into a listicle, get rounded again, and end up as a figure that appears authoritative because it appears everywhere. Tracing any given number back to a document that actually published it is often impossible. Making the provenance visible, including the gaps, is the only useful thing a publication in our position can do.
Three commitments follow.
We publish no figure without a named source. An empty cell is an honest state and we use it freely.
We do not reconcile disagreements. Where two authorities publish different figures for the same thing, both appear with both citations, and we do not average them, split them, or quietly prefer the one we like. A number that no authority published is not more useful for having been derived from two that did.
We do not fill gaps by analogy. If nothing is published for cooked fennel confit, we do not reason across from cooked vegetables generally and present the result as guidance. Reasoning by analogy is how most of the confident wrong advice in this category gets made.
How to read the table
Each row carries five things.
The item. Described specifically enough to be matched to something in your fridge, since "leftovers" is not a category anyone published a figure for.
The condition. Refrigerated, frozen, or in some other stated state. A duration without a condition is meaningless, and most of the numbers circulating online have been separated from their condition at some point in the copying.
The authority. The organisation that published the figure, named. Not "food safety experts", not "the guidelines", not a paraphrase of an agency we did not consult. If the cell does not name a specific body, we did not retrieve a specific document.
The published figure. Transcribed or tightly summarised with its conditions visible. Follow the link for the authority's full wording. Both temperature scales appear where a conversion is useful.
The retrieval date. The date we obtained it. Guidance changes. A citation without a date is a claim about the present tense that may have expired.
There is also a state marker on every row, and there are exactly three values it can take.
| State | Meaning |
|---|---|
| Cited | A named authority publishes a figure, shown with a direct link and retrieval date. |
| Authorities disagree | More than one named authority publishes a figure and the figures differ. Both are shown. Neither is preferred. |
| No published figure located | We looked and did not find a published figure from a named authority for this specific item and condition. This is a real, usable answer and it is shown rather than hidden. |
The third state is the one that makes this page different from every other storage chart, and it appears more often than readers expect.
The table
The authorities do not publish a distinct domestic storage window for every preparation readers may search for. The table therefore distinguishes specific guidance from broader categories instead of filling gaps by analogy.
| Item | Condition | Authority | Published figure | Retrieved | State |
|---|---|---|---|---|---|
| Homemade garlic or herb oil | Refrigerated | CDC botulism prevention | Refrigerate and discard unused oil after 4 days | 21 August 2026 | Cited |
| Garlic in oil | Frozen | National Center for Home Food Preservation | Up to several months | 21 August 2026 | Cited; not generalised to every herb oil |
| Cooked leftovers, including stocks, braises, cooked meat and cooked vegetables | Refrigerated | USDA FSIS | 3 to 4 days | 21 August 2026 | Cited as broad-category guidance |
| Cooked leftovers | Frozen | USDA FSIS | 3 to 4 months for quality; continuously frozen food remains safe longer | 21 August 2026 | Cited as broad-category guidance |
| Meat or vegetables stored under fat; strained cooking fat | Refrigerated or frozen | No preparation-specific consumer figure located in the FDA, USDA FSIS or NCHFP pages reviewed | — | 21 August 2026 | No published figure located |
| Raw meat in a seasoning brine | Refrigerated | FDA Safe Food Handling | FDA says to marinate in the refrigerator; no brine-specific duration is stated | 21 August 2026 | Cited handling rule; no duration located |
| Cold-cured fish | Refrigerated | FDA seafood safety | No gravlax-specific domestic duration stated | 21 August 2026 | No published figure located |
| Egg-thickened sauce, custard, homemade mayonnaise or similar preparation | Refrigerated | FDA egg safety | No preparation-specific domestic duration stated | 21 August 2026 | No published figure located |
| Food cooked in a sealed bag at low temperature | Refrigerated | No consumer storage schedule located that makes the sealed bag, by itself, a distinct safe-storage category | — | 21 August 2026 | No published figure located |
| Reheated leftovers | Refrigerated or frozen | USDA FSIS | Reheat to 165°F; USDA also describes safe refreezing after reheating | 21 August 2026 | Cited |
The permanent admission rule is that every populated row requires a cited source. Hazard-carrying posts link here rather than restating a number locally: one place to update means one place that can go stale, instead of thirty.
Where authorities disagree
Disagreement between credible bodies on storage duration is common, and it is not a scandal. Different authorities write for different populations, assume different domestic equipment, apply different safety margins, and sometimes answer subtly different questions. One may publish a conservative figure aimed at the general public; another may publish a figure aimed at a food business with monitored equipment. Both can be correct within their own frame.
What we will not do is resolve it for you. When two named authorities publish different durations for the same item and condition, the row shows both, side by side, each with its own citation and retrieval date, and the state marker reads Authorities disagree.
Averaging them would manufacture a number with no source, which is the exact failure this page exists to correct. Picking the longer one would substitute our preference for their judgement. Picking the shorter one would look like caution while still being an unattributed editorial act. Showing both is the only version where the reader can see what is actually known.
If you need a single number to act on, the conservative published figure is the one to use, and that is a decision you are making rather than one we made for you.
Where no authority publishes a figure at all
This section matters more than most readers expect, and it is the most useful part of the page.
There are a great many specific preparations for which no named authority publishes a storage duration. Not because the question is unimportant, but because published guidance is organised around broad food categories and common commercial products, and a jar of fennel confit in olive oil is not one of those. The gap is real and it does not close by being ignored.
When a row reads No published figure located, it means we searched the guidance of named authorities and did not find a figure for that specific item under that specific condition. It does not mean the food is safe indefinitely. It does not mean the food is dangerous. It means the number you were looking for does not exist in the place you would expect to find it, and anyone who gives you one confidently has either found a source we missed, in which case we would like to hear about it, or made it up.
A no-figure row is not permission to borrow a duration from a neighbouring category. Make only what you will use promptly, or ask the relevant food-safety authority for product-specific advice. Several techniques on this site are written in small batches specifically to keep that option practical.
Cooling before storing: the step that decides everything
A storage duration assumes the food got into storage properly. If it did not, the duration does not apply, and this is the part of the subject most often skipped.
The problem is physical. A large volume of hot food cools slowly, and it cools from the outside inward. The centre of a full stockpot or a deep tray of braise can remain warm for a long time after the surface feels cool, and the exterior gives you no information about the interior. Depth, volume, vessel material and whether the container is covered all change the curve.
For a domestic kitchen, USDA FSIS says to refrigerate leftovers within two hours, divide large quantities into shallow containers, and use a refrigerator at 40°F (4°C) or below. The FDA Food Code's staged cooling limits are written for regulated food establishments; they are useful context, but this page does not silently recast a retail code as domestic law.
Three structural points that require no number.
Volume is the variable you control. Dividing a large batch into shallower containers changes the cooling curve more than anything else available to a home kitchen, because it changes the distance heat has to travel.
Surface temperature is not interior temperature. Touching the side of a pot tells you about the pot. A probe in the centre tells you about the food.
The hob is not a cooling method. Leaving a covered pot to cool slowly on the hob overnight is the specific practice that makes the cooling question urgent, and it is addressed at length in what resting does over twenty-four hours. Large-batch stock has the same problem and is covered in stock below the simmer.
Sensory checks are not a safety test
This needs stating plainly because the habit is so widespread.
Smelling food, looking at it, and tasting a small amount are useful for detecting spoilage of the ordinary kind, where organisms produce odours, slime, gas and colour changes as they multiply. Those signals are real and worth heeding.
They are not a safety test. Several of the hazards that matter most in low-temperature cooking produce no detectable change at all. Food carrying them can look, smell and taste exactly like food that does not. A sniff that finds nothing has told you that you found nothing, which is not the same as there being nothing.
So the sequence is: apply the cited storage window; if the food is outside it, discard it regardless of how it seems. Sensory judgement can move you toward discarding something early. It cannot be used to extend a window.
Where to go for guidance we cannot give
Some questions are outside anything this page will ever cover.
Home canning and shelf-stable preserving. Permanently out of scope on this site. The CDC directs home canners to the tested processes in the USDA Complete Guide to Home Canning and warns against substituting other recipes. Use those procedures directly.
Anything specific to your circumstances. Guidance written for a general population is exactly that, and questions about particular circumstances belong to a qualified professional rather than to a cooking publication.
Anything where you cannot find a published figure and cannot accept the risk of guessing. Contact your national food-safety authority. They answer questions from the public, and it is a better use of ten minutes than three hours of contradictory search results.
In Plain English
What it means. This page collects what named authorities publish about how long food keeps, names them, dates the retrieval, and leaves the gaps visibly empty. The numbers are theirs. The indexing is ours.
Who it helps. Anyone standing in front of an open fridge trying to work out whether a search result from an unattributed blog post is a real number. Usually it is not.
When to use it. Before you store something, not after. And before you rely on a figure you half-remember, since half-remembered storage numbers are how most of the bad ones spread.
What to do next. Find your item and its condition in the table. If the row is cited, follow the authority, not us, and check the retrieval date. If the row shows a disagreement, take the conservative figure and know that you chose it. If the row says no published figure was located, take that seriously and consider making the thing in a quantity you will finish. And whichever row you land on, read the cooling section, because a storage window applied to food that was cooled badly is a number describing a situation you are not in.
The next thing to read
Storage duration is one half of a subject whose other half is cooling, and neither works without the other.
Read the authorities' own pages linked in the table, which carry the cooling, holding and reheating guidance the table depends on, and check each against its retrieval date.
Related technique: When The Variable Is Hours, Not Degrees.