How to Check CPU Temperature on Windows
How to check CPU temperature on Windows, the short version: Windows has no built-in reading, so you need a monitoring tool that shows the sensor Windows itself does not expose. This page covers the fastest way to see a live reading, what counts as normal, how to check under real load rather than just idle, and what to do if the number looks concerning.
- No Windows built-in
- Live, per core
How to check CPU temperature in under a minute
Windows does not show CPU temperature anywhere in Task Manager, Settings, or any built-in utility — which surprises a lot of people the first time they go looking for it. The fastest reliable way to see how to check CPU temperature on a running system is a lightweight monitoring tool that reads the sensor directly from the CPU and displays it live.
- 1
Download and install Quick CPU
Grab the installer from the download page — free, no account, and the dashboard shows live readings within seconds of opening.
- 2
Open the main dashboard
Per-core temperature is one of the headline readings on the main screen, updating live with no configuration needed.
- 3
Check the per-core sensor view for detail
For a full breakdown — every core, current/min/max/average, and distance to Tj Max — open the dedicated sensor view rather than reading the single summary number.
- 4
Load the CPU if you want a worst-case reading
Idle temperature tells you little about thermal headroom. Run the built-in stress test to see how to check CPU temperature under sustained full load, which is the reading that actually matters for stability.
Why is there no built-in way to check CPU temperature?
Temperature is reported by a sensor built into the CPU die itself, read through a low-level interface Windows does not surface in its own interface by default. Task Manager shows utilisation and, on some systems, clock speed, but never the sensor reading that answers how to check CPU temperature directly. That gap is exactly why third-party monitoring tools exist and remain widely used even though CPU temperature is arguably as fundamental a reading as utilisation.
The one native exception is the BIOS/UEFI setup screen most motherboards show at boot, which typically displays an idle temperature reading before Windows loads. It answers how to check CPU temperature in the narrowest possible sense — a single idle number, requiring a reboot to see, with no way to watch it change under real load, and no per-core breakdown to speak of either.
What temperature should you actually expect?
There is no single universal safe number, because it depends on your specific CPU model and cooling setup, but rough expectations hold across most modern systems: idle temperatures in the 30-50°C range, and load temperatures anywhere from the 60s into the 80s°C with adequate cooling. Laptops and small-form-factor systems commonly run warmer than desktop towers under the same load, which is normal given their more constrained cooling, not a sign of a problem on its own.
The more useful figure than a raw number is distance to your CPU's maximum rated temperature — often labelled Tj Max — since that tells you how much headroom actually remains rather than comparing your reading to a generic range that may not apply to your specific chip. Intel's own guidance on processor temperature covers this distinction in more depth for Intel chips specifically. A good answer to how to check CPU temperature shows that distance directly rather than leaving you to look up your CPU's specification sheet separately.
How to check CPU temperature under real load
An idle reading tells you the system is not obviously broken, but it does not answer whether your cooling can actually keep up when the CPU is working hard — which is the situation that actually matters for stability and longevity. The reliable way to check is to load the CPU deliberately, watch the live reading while it runs, and see where it settles rather than how high it spikes momentarily.
A dedicated stress test is the clean way to do this: sustained, predictable load with a timer, rather than trying to keep a game running long enough to judge thermal behaviour informally. Watching per-core temperature during that test — not just one blended average — is how to check CPU temperature in a way that would actually catch one core running hotter than the rest, which a single overall number can hide.
What to do if your CPU temperature looks concerning
First, put the number in context: compare it against distance to your specific CPU's thermal limit, not a generic range that might belong to a different chip entirely. If it is genuinely close to that limit, the usual physical causes are worth checking first — a cooler that is not properly seated, dust buildup restricting airflow, or a case with inadequate ventilation for the hardware inside it.
If the physical setup checks out, software settings can also be a factor: aggressive core parking behaviour or a power plan pushing more sustained load onto more cores than your cooling was designed for can both raise sustained temperature. See our core parking guide and power plan guide for adjusting either, and deliberately capping clock modulation or turbo power limits is a direct, reliable way to bring temperature down if a hardware fix is not immediately available.
It is also worth ruling out ambient conditions before assuming something is wrong with the hardware itself: a warm room, a laptop placed on a soft surface that blocks intake vents, or simply the middle of summer without air conditioning can all shift readings noticeably higher than they would be under cooler ambient conditions, without any change to the CPU or cooler at all. Comparing a reading taken on a hot afternoon against one from a cool morning is not really how to check CPU temperature fairly — try to compare readings taken under similar ambient conditions when judging whether something has actually changed.
A high temperature reading and a high utilisation reading are related but not the same question — this page covers reading the temperature itself; whether sustained high usage is a problem in the first place, including what to check if nothing you're running seems to explain it, is covered in is 100% CPU usage safe.
Mistakes that make a temperature check misleading
A few habits quietly undermine an otherwise correct method for how to check CPU temperature. Checking only at idle and assuming load behaves the same way is the most common — idle and load readings can differ by 30-40°C or more on the same system, so an idle-only check answers a different question than the one that usually prompted the search in the first place.
Reading a single instantaneous number rather than watching it over a few minutes is another: CPUs spike and settle constantly under real workloads, and catching one high reading in passing tells you far less than watching the maximum and average over a sustained period. A monitoring view with live minimum, maximum and average columns solves this directly — it is built to answer how to check CPU temperature as a pattern over time, not as a single lucky or unlucky snapshot.
Comparing your reading against a number you saw online for a "similar" CPU is the third common mistake — cooling varies enormously between systems even with the identical chip, and a temperature that is completely normal on a well-ventilated desktop can be a genuine concern on a thin laptop with the same processor. Your own distance to your specific CPU's thermal limit is the only comparison that is actually meaningful.
Why a one-time check isn't the whole answer
A single reading, however carefully taken, only describes one moment. Thermal paste degrades over months and years, dust accumulates gradually, and a system that ran cool when new can creep upward without any single dramatic change to point to. Revisiting how to check CPU temperature periodically — not just once when something feels wrong — is what actually catches that kind of slow drift before it becomes a real problem.
Keeping a live dashboard open occasionally during normal use, rather than only running a one-off check, is the simplest way to build a rough mental baseline for what your specific system normally reads — which makes it much easier to notice when a reading is genuinely unusual for your machine, rather than guessing against a generic range that was never really about your hardware in the first place.
How to check CPU temperature: quick answers
How do I check CPU temperature on Windows without extra software?
Most motherboard BIOS/UEFI screens show CPU temperature at boot, before Windows even loads — useful for a rough idle check, but you have to reboot to see it and it will not show temperature under real load. Windows itself has no built-in temperature reading in Task Manager or Settings, which is why third-party software is the practical answer for how to check CPU temperature while actually using the machine.
What is a normal CPU temperature?
It depends on your specific CPU and cooling setup, not on a single universal number — see the ranges below for the rough idle/load figures most desktops and laptops fall into. What varies most between two otherwise-identical CPUs is the cooling around them: a thin laptop runs warmer than a desktop tower under the same load, and that alone can be a 15-20°C difference without either machine having a problem. Distance to your own CPU's maximum rated temperature (Tj Max) is the number that actually answers whether a given reading is fine.
How to check CPU temperature under load, not just idle?
Idle readings are reassuring but do not answer the question that usually matters — whether your cooling can keep up when the CPU is actually working. Running a stress test while watching the live temperature reading is how to check CPU temperature at its realistic worst case; see our stress test guide for running one safely.
My CPU temperature looks high — what should I do?
First confirm it is genuinely high relative to your specific CPU's limit, not just a number that sounds high in isolation — distance to Tj Max is the more useful figure than the raw reading. If it is genuinely close to the limit, check that your cooler is properly seated and dust-free, that your case has adequate airflow, and consider whether core parking or power plan settings are keeping more cores active than your cooling can handle.
Does Quick CPU show every core separately when checking temperature?
Yes — the per-core sensor view lists temperature for each individual core, not just one blended figure, along with current, minimum, maximum and average values and each core's distance to its own thermal limit. That per-core detail is often how to check CPU temperature usefully rather than superficially, since an average can hide one core running much hotter than the rest.
Read next
See your own temperature, live and per core
Free, no account — the fastest way to check CPU temperature on your own machine, updated in real time.