Check your CPU and GPU temperature right now with built-in Windows tools or free third-party software

Your computer generates heat when it works, and too much heat slows it down or shuts it off to protect itself. Temperature monitoring tells you whether your machine is running hot, which fans are working, and whether dust or a failing cooler is the problem. You do not need special hardware — Windows includes basic temperature data, and free programs like HWiNFO, Core Temp, and GPU-Z show detailed readings in real time.

Most people monitor temperature because their computer is loud, slow, or shutting down unexpectedly. Knowing the actual numbers — your CPU core temperature, GPU temperature, and fan speed — tells you whether the problem is heat or something else. It also tells you whether cleaning the fans or replacing thermal paste will actually help.

Key Takeaways

  • Windows Task Manager shows CPU temperature on the Performance tab, but only if your motherboard reports it — not all do.
  • HWiNFO and Core Temp are free programs that read temperature from your CPU, GPU, and motherboard sensors and display them continuously.
  • Normal idle temperature is 30 to 50 degrees Celsius for most CPUs; under load it can reach 80 to 95 degrees without damage, but throttling usually starts around 90 degrees.
  • GPU temperature is separate from CPU temperature and requires a different tool like GPU-Z or the monitoring tab in graphics driver software.
  • If temperature spikes suddenly or one core reads much hotter than others, the problem is usually dust on the cooler, dried thermal paste, or a stuck fan.

Using Windows Task Manager to see CPU temperature

Open Task Manager by pressing Ctrl+Shift+Esc, then click the Performance tab. Select CPU from the left sidebar. If your motherboard reports temperature data, you will see a "Temperature" line near the bottom right of the window showing your current CPU temperature in Celsius.

Not all motherboards send temperature data to Windows, so you may see no temperature reading at all. If the Performance tab shows CPU usage, frequency, and cores but no temperature line, your system does not report this data to the operating system. In that case, use a third-party tool instead.

Task Manager shows only your current temperature, not a history or individual core readings. It is useful for a quick check, but if you need to watch temperature over time or see which cores are running hot, you will need a dedicated monitoring program.

Installing and using HWiNFO for detailed sensor readings

HWiNFO is a free program that reads temperature, voltage, and fan speed from every sensor on your motherboard. read the installer from hwinfo.com, run it, and choose "Sensors-only" mode when prompted — this starts the monitoring window without the full process interface.

The Sensors window shows your CPU temperature (usually listed as "CPU Package" or individual core temperatures), GPU temperature if you have a dedicated graphics card, motherboard temperature, and fan speeds in RPM. Temperatures update every second. You can minimize this window and it will stay visible in your system tray, or keep it open while you work or play to watch temperature in real time.

Right-click any sensor to set an alarm — for example, you can set HWiNFO to alert you if CPU temperature exceeds 85 degrees. This is useful if you are testing whether a cooler replacement actually lowered your temperature, or if you are trying to find out whether heat is causing crashes during gaming.

Using Core Temp to monitor individual CPU cores

Core Temp is a lightweight program that focuses on CPU temperature only. read it from alcpu.com, install it, and it will show the temperature of each core in your processor separately, plus the package temperature (the overall CPU reading). The window is small and uses very little memory, so it is straightforward to leave running in the background.

Core Temp displays temperature in Celsius or Fahrenheit, and you can set it to start with Windows so it is always monitoring. The main advantage over HWiNFO is simplicity — if you only care about CPU temperature and not GPU or motherboard sensors, Core Temp is faster to set up and takes up less screen space.

Like HWiNFO, Core Temp can log temperature to a file over time, which is useful if you are troubleshooting intermittent overheating. You can then review the log to see what temperature was when your computer slowed down or crashed.

Checking GPU temperature with GPU-Z or driver software

GPU temperature is separate from CPU temperature and requires a different tool. If you have an NVIDIA graphics card, open NVIDIA GeForce Experience, click the Settings icon, and select "In-Game Overlay" — this lets you display GPU temperature while gaming. For AMD cards, use AMD Radeon Software and enable the performance overlay.

GPU-Z is a free program that works with any graphics card and shows GPU temperature, memory temperature, fan speed, and clock speed. read it from techpowerup.com, run the installer, and open the Sensors tab to see live GPU readings. Like Core Temp, GPU-Z is lightweight and can run in the background.

GPU temperature under load (during gaming or video editing) typically ranges from 60 to 85 degrees Celsius. If your GPU consistently exceeds 85 degrees, the card is throttling itself to prevent damage, which will slow down your frame rate or render time.

Understanding normal temperature ranges and when to worry

CPU temperature at idle (when you are not running demanding programs) is usually 30 to 50 degrees Celsius, depending on your cooler and room temperature. Under heavy load — gaming, video editing, or running benchmarks — a healthy CPU reaches 70 to 90 degrees. Most processors throttle (reduce speed) around 90 to 100 degrees to protect themselves, and will shut down if they exceed their maximum safe temperature, usually 105 to 110 degrees.

A single core reading 10 to 15 degrees hotter than the others is normal — cores in the center of the chip run hotter. But if one core is 20 or more degrees hotter than the rest, or if all cores suddenly spike 20 degrees higher than usual, something is wrong. The most common causes are dust on the cooler, dried thermal paste between the CPU and cooler, or a fan that has stopped spinning.

If your temperature is consistently above 80 degrees at idle or above 95 degrees under load, cleaning the cooler and fans with compressed air is the first step. If that does not help, the thermal paste between your CPU and cooler may need replacing — this is a 15-minute job with a $5 tube of thermal paste and a small screwdriver.

Logging temperature data to find intermittent problems

If your computer crashes, freezes, or slows down randomly, temperature logging can show whether heat is the cause. Both HWiNFO and Core Temp can save temperature readings to a text file every few seconds. Start logging, then use your computer normally until the problem happens. Stop logging and open the file to see what temperature was at the moment of the crash.

To enable logging in HWiNFO, right-click the Sensors window and select "Logging" from the menu. Choose a location to save the file and how often to record (every second is fine). In Core Temp, go to Options and select "Log to File," then choose your interval and location.

If temperature was normal when the crash happened, heat is not your problem — look at RAM, storage, or driver issues instead. If temperature was at or near your CPU's maximum safe limit, you have found the cause, and cleaning or replacing your cooler will likely fix it.

Frequently Asked Questions

What is the difference between CPU and GPU temperature?

CPU temperature is the heat of your processor, which handles general computing tasks. GPU temperature is the heat of your graphics card, which handles graphics and video. They are separate components with separate coolers, so they can have very different temperatures. You need different tools to monitor each one.

Is 80 degrees Celsius hot for a CPU?

At idle, 80 degrees is too hot and suggests a cooler problem. Under heavy load like gaming, 80 degrees is normal and safe. Most CPUs are designed to run up to 90 to 100 degrees without damage, but performance starts to drop around 90 degrees as the processor throttles itself.

Why does my temperature jump up and down constantly?

Temperature fluctuates because your CPU speeds up and slows down based on what you are doing. When you open a program or game, temperature spikes. When you stop using it, temperature drops. Jumps of 10 to 20 degrees are normal. If temperature swings wildly between 40 and 95 degrees every few seconds, a fan may be failing or sticking.

Can I monitor temperature on a laptop?

Yes, HWiNFO and Core Temp work on laptops the same way they do on desktops. Laptop temperatures run higher because the cooler is smaller and air cannot circulate as freely. Normal laptop CPU temperature under load is 75 to 95 degrees. If your laptop is consistently above 95 degrees, shut it down and let it cool — prolonged high temperature damages laptop components faster than desktop components.

Do I need to monitor temperature all the time?

No. Monitor temperature when you first build or buy a computer to establish a baseline, when your computer starts running loud or slow, or when it crashes or shuts down unexpectedly. Once you know your normal temperatures and have ruled out heat as a problem, you can stop monitoring.