Guide
Safe internal cooking temperatures for meat: the USDA chart and where chefs disagree
The USDA chart is a single instant-kill temperature. Real food safety is time + temperature — and chefs know it.
By Buğra SözeriPublished Updated
The USDA publishes a chart of safe minimum internal temperatures for meat. The chart is widely cited but widely misunderstood — the numbers are instant-killtemperatures, the point at which pathogens die almost immediately. Pathogens also die at lower temperatures, just more slowly. This is why sous-vide cooking at 130°F produces safe medium-rare beef despite being below the USDA chart’s 145°F figure.
The USDA chart (2026)
| Food | Safe min internal temp (°F) | °C | Rest time |
|---|---|---|---|
| Whole cuts of beef, pork, lamb, veal (steaks, roasts, chops) | 145 | 63 | 3 min |
| Ground meat (beef, pork, lamb, veal) | 160 | 71 | — |
| Poultry (whole, ground, parts, stuffing) | 165 | 74 | — |
| Egg dishes | 160 | 71 | — |
| Fish and shellfish | 145 | 63 | — |
| Leftovers and casseroles (reheating) | 165 | 74 | — |
| Ham (fresh or precooked, reheating) | 145 / 165 | 63 / 74 | — |
Why ground meat needs higher temps than whole cuts
On a whole steak, pathogens live on the surface only. The moment you sear it, surface bacteria die; the interior is irrelevant because it was never exposed to surface contamination.
On ground meat, surface bacteria are mixed throughout the product during grinding. Every spot of the patty has the same pathogen risk as the original surface. So you have to bring the interior to a temperature that kills pathogens throughout — hence 160°F, well above the 145°F a steak needs.
Exception: whole-muscle beef hand-ground at home or by a butcher you trust (where you saw the steak go in) is safer than mass-distributed ground beef where one contaminated carcass can contaminate thousands of pounds.
The time-temperature trade-off
Pasteurisation isn’t a single temperature; it’s a temperature-time curve. Salmonella, for instance, is killed:
- Instantly at 165°F (74°C)
- In about 10 seconds at 160°F (71°C)
- In about 9 minutes at 145°F (63°C)
- In about 30 minutes at 135°F (57°C)
- In about 4 hours at 130°F (54°C)
This is the basis of sous-vide cooking. A steak held at 130°F for 4 hours is just as bacteriologically safe as a steak briefly hit at 165°F — and dramatically more tender. The USDA chart is conservative because it doesn’t encode the time-axis.
Where chefs cook lower than the USDA chart
Three classic disagreements:
Beef
USDA: 145°F (medium). Steakhouse medium-rare: 130-135°F. For whole muscle (steak, roast), the surface is the only contaminated area; sear it and the inside can be much lower temperature without risk. Restaurant medium-rare is absolutely safe by sous-vide-pasteurisation standards.
Pork
USDA dropped pork from 160°F to 145°F in 2011, formally acknowledging that the old higher number was conservative. Modern pork is essentially trichinella-free in the commercial supply (US incidence: ~3-4 cases per year nationally, mostly from wild game, not farm pork). Most chefs cook pork to 140-145°F for a slight blush of pink.
Eggs
USDA: 160°F. Many chefs make eggs at 145-150°F (custards, carbonara, hollandaise) because the eggs have been pasteurised in shell or via long-hold time. Restaurant Caesar dressing, runny yolks on burgers, and similar preparations are common and safe for healthy adults but riskier for immunocompromised diners.
Where the chart is not negotiable
- Poultry. Salmonella prevalence in chicken is high (~25% of grocery-store chicken samples in CDC data). 165°F instant-kill is sensible. Sous-vide chicken at 145°F for 90 minutes works for chefs but requires confidence in temperature and timing.
- Ground meat.See above — surface contamination is mixed throughout. Don’t cook ground beef rare unless you ground it yourself from known-safe whole muscle.
- Reheated leftovers. Bacteria can multiply if food has been improperly cooled. Reheating to 165°F kills anything that may have grown.
- Pregnant, immunocompromised, very young, very old.Stick to the chart. The time-temperature equivalence works statistically but a single failed step is more dangerous to vulnerable diners.
How to actually measure
Buy a thermocouple-style instant-read thermometer (Thermapen and similar). Insert into the thickest part of the meat, avoiding bone. Read in 2-4 seconds. Cheaper bimetal-coil thermometers take 15-30 seconds and lose 5-10°F of accuracy in the process.
For roasts and turkeys, use a leave-in probe attached to a digital readout outside the oven. Set alarm at the target temperature. Pull the meat at 5°F below target to account for carryover cooking during the rest period.
The pragmatic bottom line
- Poultry, ground meat, leftovers: follow the chart.
- Whole-muscle beef and pork: cook to your preferred doneness; surface contamination is killed by the sear.
- Sous-vide: time-temperature equivalence is real. Use a published table from a reputable source (Modernist Cuisine, Douglas Baldwin’s sous-vide guide).
Convert between °F and °C using our oven temperature converter. For why colour is unreliable as a doneness indicator (especially in pork and chicken), see the cooking methodology page.
Worked example: a 2-inch ribeye, sous-vide vs grill
A 2-inch thick (~5 cm) bone-in ribeye, 600 g, refrigerator- cold at 4 °C (40 °F).
- USDA pan-sear path: sear 4 min/side over high heat, finish in a 200 °C (400 °F) oven, pull at internal 60 °C (140 °F), rest 10 min for carryover to 63 °C (145 °F). Total time: ~20 min. Internal gradient: well-done outside (75 °C+), medium-rare core. Per USDA chart plus 3-minute rest, this is safe for healthy adults.
- Sous-vide path:seal in a bag, immerse in a 54.4 °C (130 °F) water bath for 2 hours (Baldwin pasteurisation table for ≥6.5-log Salmonella reduction on a 50 mm thickness). Sear 60 sec/side over screaming-hot cast iron. Total active time: ~5 min, total elapsed: ~2 h 5 min. Result: edge-to-edge medium-rare with a Maillard crust. Pasteurised by time-temperature equivalence; safe for healthy adults (not pregnancy / immunocompromised).
- Where they differ: texture (sous-vide is more tender by 15-20% per WBSF shear-force tests), waste (sous-vide is 0% gradient cook loss vs 12-18% on the oven path), risk profile (sous-vide depends on bath accuracy ±0.5 °C; oven path tolerates ±5 °C of error).
Common mistakes
- Trusting the “poke test” or colour.Doneness colour depends on the meat’s myoglobin content (chicken thigh stays pink at 80 °C; pork loin grays at 60 °C). USDA explicitly states colour is not a reliable doneness indicator. Use a calibrated probe.
- Probing too shallow. The geometric centre of a roast or large bird is the slowest-warming point. A probe inserted 2 cm from the surface reads 10-15 °C hotter than the centre on a thick joint. Insert to the centre.
- Skipping the rest.The USDA rest time on whole cuts (3 minutes at ≥63 °C) is part of the pasteurisation — internal temperature continues climbing and the time-at-temperature finishes the kill.
- Cross-contamination. The thermometer probe used in raw chicken must be cleaned with hot soapy water before re-use. CDC traces a meaningful fraction of home Salmonella outbreaks to this single oversight.
- Stuffing in poultry.The stuffing inside a turkey reaches 74 °C (165 °F) much slower than the bird’s thigh. USDA recommends cooking stuffing in a separate dish for exactly this reason.
Edge cases the USDA chart doesn’t fully cover
- Game meat.Wild boar carries genuine trichinella risk (commercial pork doesn’t). Cook game pork to 71 °C (160 °F), not 63 °C. Venison from a known source can follow whole-muscle beef rules; wild game from unknown sources should follow ground-meat rules.
- Mechanically tenderised steaks.“Blade tenderised” or “needle tenderised” steaks have surface bacteria pushed into the interior, just like ground meat. Cook to 71 °C (160 °F) regardless of cut. USDA mandates labelling these as of 2016.
- Reheating commercial cured meats.Ham labelled “ready to eat” can be served cold; if reheated, 60 °C (140 °F) is enough. Ham labelled “cook before eating” needs 63 °C with a 3-min rest.
- Vacuum-sealed meat held warm. Sous-vide bags held below 54.4 °C / 130 °F enter the danger zone for C. botulinumspore germination if held >4 hours. Keep the bath above 54.4 °C or chill rapidly.
- Microwave cooking. Uneven heating creates cold spots that test thermometers miss. USDA requires rotating, stirring, and a 2-min covered rest, with multi- point thermometry.
Sources: USDA Food Safety and Inspection Service, “Safe Minimum Internal Temperatures” (2024 ed.); Douglas Baldwin, A Practical Guide to Sous Vide Cooking; FDA Food Code 2022; CDC Salmonella in Poultry surveillance reports 2023.
Frequently asked questions
- What is the USDA safe internal temperature for chicken?
- 165°F (74°C) for all poultry — whole birds, ground chicken, and parts. No rest time is required at this temperature. Salmonella is present in roughly 25% of grocery-store chicken samples, so the 165°F instant-kill threshold is not conservative — it is appropriate.
- Can I cook a beef steak to medium-rare and still be food safe?
- Yes for whole-muscle cuts. Pathogens live only on the surface of an intact steak; searing kills them before the interior temperature matters. Restaurant medium-rare (130–135°F interior) is safe for healthy adults. The USDA's 145°F is a conservative instant-kill standard, not the only safe endpoint.
- Why does ground beef need to be cooked to 160°F when steaks only need 145°F?
- Grinding mixes surface bacteria throughout the entire product. In a steak, pathogens are only on the surface and are killed by searing. In ground beef, every interior particle carries the same risk as the original surface, requiring the interior to reach a full kill temperature of 160°F.
- What is sous-vide cooking and is it food safe?
- Sous-vide holds food at a precise low temperature for an extended time, achieving pasteurization through time-temperature equivalence rather than high heat. A steak at 130°F for 4 hours achieves the same pathogen reduction as a brief 165°F sear. This is based on established food science — but requires accurate thermometry and is not recommended for pregnant, elderly, or immunocompromised individuals.
- What thermometer should I use to check meat temperature?
- A thermocouple-style instant-read thermometer (such as a Thermapen) reads in 2–4 seconds and is accurate to ±0.5°F. Cheaper bimetal-coil thermometers take 15–30 seconds and lose 5–10°F of accuracy. For roasts and turkeys, use a leave-in probe that alarms at the target temperature.
- Are the USDA temperature guidelines safe for everyone?
- The USDA instant-kill temperatures are appropriate for healthy adults. Pregnant people, infants, adults over 65, and the immunocompromised should follow USDA guidelines strictly and avoid any preparation that relies on time-temperature equivalence below the listed minimums. When in doubt, follow the official chart.
Sources & references
Authoritative references cited by this piece. Verified by Buğra Sözeri on the dates shown and re-checked at every deploy.
- USDA Food Safety and Inspection Service — Safe Minimum Internal Temperatures — Authoritative US chart of instant-kill minimum temperatures referenced throughout the article(as of )
- FDA Food Code 2022 — Time-Temperature Pathogen Control — Regulatory reference for the time-temperature pasteurisation tables (the sous-vide-friendly numbers)(as of )
- Modernist Cuisine — Pasteurisation Reference Tables — Engineering-grade reference for the equivalent-pasteurisation time-temperature curves cited(as of )
- Douglas Baldwin — A Practical Guide to Sous Vide Cooking — Peer-cited engineering reference for the time-temperature pasteurisation tables used in the sous-vide section(as of )
- CDC — Food Safety: Foodborne Illness Surveillance — Authoritative US source for foodborne pathogen prevalence and outbreak data referenced in the article(as of )
Related
Published May 16, 2026 · Last reviewed May 31, 2026