PowerFlow vs MacBookBatteryMonitor: per-app watts that add up

What PowerFlow does

PowerFlow (dylansantwani/powerflow on GitHub) is an open-source menu bar app that draws the Mac's power as a Sankey diagram from adapter and battery to system load, with a runtime-if-you-unplug figure and a "Draw by app" list. All 6 of its commits landed on 6 September 2026, it has no releases, and you install it by building from source with ./build.sh, which produces an ad-hoc signed app. The app list reads each process's kernel CPU energy counter every 3 seconds and folds helpers into their parent app. The README states what that covers: the counter "accounts for CPU energy, not the whole machine". Under a controlled load, system power rose 20.8 W while app energy rose 9.8 W, and at idle the apps account for 10 to 25% of system draw. WindowServer cannot be read at all, root-owned daemons are skipped, and the display, the GPU, and everything else land in a "System & display" row that the README calls usually the largest number on screen. In the README's own screenshot it reads 94 W of a 104 W load. PowerFlow is live only, with no history. Its draggable markers notify you when the charge crosses a level and do not change how the Mac charges, and it has no per-app controls. The older, better-known PowerFlow (lzt1008/powerflow, 364 stars as of October 2026) has not been updated since 29 March 2025, has open crash reports on macOS 26 and 27, has open reports that screen power stays at 0.0 W, and shows no per-app watts at all.

What MacBookBatteryMonitor does

MacBookBatteryMonitor leaves nothing in an unexplained bucket. It measures your whole Mac's power every second and breaks it into 16 parts (CPU, GPU, Neural Engine, memory, display, SSD, network, fans, USB and more) that add up to the total in its Calibrated view, then splits it between your apps by how much work each one does, so every app's share adds up to what your Mac actually drew. Its Raw Sensors view shows every sensor exactly as reported, plus a "Not Covered" line that shows what chip-only tools miss. Its adaptive algorithm tunes itself to your exact Mac in the first 15 minutes, and from then on its power readings are 99.63% accurate across our lab runs on an M1 Max MacBook Pro and an M2 Mac mini, checked against power measured at the battery and at the device. Helpers roll up into their app, short-lived build and test workers are counted, and Docker, cloudflared, and launchd services are tied to the process hosting them. Per-app charts run from 1 hour to 7 days, and 30 days of every-process data stay on the Mac, on battery and on the charger. It also acts: cap any app's CPU from 10% to 100%, push it onto the efficiency cores, apply the rule always, only on battery, or below a battery level you pick, or quit a runaway app in one click. A 50% to 100% charge limit with Sailing Mode controls the battery itself. It is a download, not a build, and as of October 2026 costs $9 once with lifetime updates.

Which one should you pick?

PowerFlow's Sankey diagram is a clear way to watch adapter and battery watts, and it is free if you are willing to build it yourself. For per-app power it leaves most of the draw unexplained: by its own measurements, apps cover 10 to 25% of system draw at idle. Its CPU-only counter sees less than chip readings like Apple's powermetrics, which saw 26.06% of the energy a 16-inch M1 Max MacBook Pro used in a real working session, and 44.81% across all our lab runs. MacBookBatteryMonitor saw 99.52% on that MacBook Pro, and its power readings are 99.63% accurate across every run. Add per-app history charts, 30 days of every-process data, per-app power limits, and a charge limit, in an app you download and install, and MacBookBatteryMonitor is the better tool, at $9 as of October 2026.

Related guides

• MacBook battery usage by app
• Best Mac power monitor apps