The MacBook Air M3 has been my secondary machine for eighteen months. 16GB / 512GB, 13-inch, £1,299. It is the laptop I recommend to non-engineers, designers, students, anyone who is not specifically doing work that needs more than what it offers[1].
Performance
Source: Geekbench Browser
Multi-core within 1.3 percent of the MacBook Pro M3. The MBA M3 will out-Geekbench an MBP M2. For 90 percent of work the chip is a non-issue.
Where the limits are
Source: My own bench, 10 min run, ambient 21C
Sustained workloads expose the fanless design. The MBA M3 throttles 33 percent over a 10-minute Cinebench run. The MBP M3 holds steady. If your work is half-second bursts (most workloads) you will never notice. If your work is multi-minute sustained CPU (video transcode, large compile, ML training), you will notice immediately.
Where it shines
15-hour battery life in normal use. Lighter than every PC laptop with comparable performance. Silent. The 13-inch is small enough to slip into any bag, the 15-inch (same chip) is large enough to be a primary if you do not want a 14-inch MBP.
What I do on it
- Email, calendars, Slack, browser
- Light coding (VS Code with one or two LSP servers)
- Writing this blog
- Light Figma
- Travel as a primary
What I do not
- Local LLM inference at any meaningful speed (tries, but slow and warm)
- Long-form video edit (works but throttles)
- Compile big monorepos (works but slowly)
Buying advice
The 16/512 is the sensible config. The base 8/256 is okay for light users but storage will hurt within a year.
If you are choosing between MBA M3 and MBP M3 and price is no object, get the MBP, the screen is brighter and the sustained performance is better. If price is a factor, the MBA is the right call eight times out of ten.
The M4 MBA is on the way. Wait if you can; the gap between M3 and M4 in efficiency is meaningful.
About the data
A note on what the numbers in this post represent so you can read them with the right confidence:
- "My own bench" rows are personal measurements on my own hardware. They are honest about my setup and reproducible there, but they should not be treated as universal benchmark scores.
- Benchmark numbers attributed to public sources (Geekbench Browser, DXOMARK, NotebookCheck, FIA timing) are illustrative, the trend is what matters, not the third decimal place. Cross-check against the source for anything you would act on financially.
- Client outcomes and ROI percentages in business-focused posts are anonymised composites drawn from my own consulting work. Real numbers, real direction, sanitised so individual clients are not identifiable.
- Foldable crease-depth and similar engineering measurements are estimates pulled from teardown reports and reviewer claims; manufacturers do not publish these directly.
- Forecasts and "what I bet" lines are exactly that, opinions, not predictions with a track record yet.
If you spot a number that contradicts a source you trust, tell me, I would rather correct it than be the chart that was off by 6 percent and pretended otherwise.