Laptops

For running AI models locally, memory matters more than anything else, because the whole model has to fit in it. For the biggest models, the MacBook Pro with M5 Max leads, with up to 128GB of unified memory the GPU can use. On Windows, a laptop with an RTX 5090 (24GB of video memory) or RTX 5080 (16GB), such as the Lenovo Legion Pro 7i Gen 10, is the fastest for models that fit. The ASUS ProArt PX13 brings 128GB of shared memory to Windows.

Check MacBook Pro M5 Max models on Amazon →

How we put this together: this guide is based on manufacturers’ specifications, NVIDIA’s and LM Studio’s published requirements, and published reviews (linked below). It is not based on hands-on testing by iTech Level, and the model-size figures are rough rules of thumb, not benchmarks. See how we choose products.

How much memory do you need to run an LLM locally?

Apps like LM Studio and Ollama load the model into memory: the graphics card’s video memory (VRAM) on a Windows gaming laptop, or the shared unified memory on a Mac or an AMD Ryzen AI Max laptop. If the model doesn’t fit, it either won’t load or spills into slower system memory. Most people run “quantized” models compressed to around 4 bits per weight, which as a rough rule needs about half a gigabyte per billion parameters, plus a few gigabytes for the conversation context:

Two more things matter. Memory bandwidth sets how fast a model answers once it’s loaded, which is why an NVIDIA RTX card with its own fast GDDR7 memory is quick for models that fit. And software support: LM Studio supports Apple silicon Macs and Windows PCs, recommending 16GB+ of RAM and at least 4GB of dedicated VRAM on Windows (LM Studio).

Best laptops for local LLMs compared

The best laptops for running LLMs locally in 2026

1. MacBook Pro with M5 Max: best for large models

Apple’s unified memory lets the GPU use most of the system’s memory, and the M5 Max supports up to 128GB of it, with 460–614GB/s of bandwidth depending on the GPU configuration (Apple). That’s far more memory for a model than any laptop graphics card, so 70B-class models that won’t fit on a 24GB RTX 5090 can run here. It’s also quiet and lasts much longer on battery than a gaming laptop, and the M5 Max models start with 2TB of storage, enough for several large models. Good to know: memory can’t be upgraded, so buy the amount you’ll need, and an RTX 5090 is faster for models small enough to fit in its 24GB.

Check MacBook Pro M5 Max models on Amazon →

2. Lenovo Legion Pro 7i Gen 10: fastest on Windows

NVIDIA’s RTX 5090 laptop GPU has 24GB of GDDR7 and the RTX 5080 has 16GB (NVIDIA), and the Legion Pro 7i offers both. It runs the RTX 5080 at the full 175 W next to an Intel Core Ultra 9 275HX; Notebookcheck rates it 88% and calls the Gen 10 the most refined model in the series (Notebookcheck). The RTX 5090 version costs considerably more; Notebookcheck lists it at about €4,100 (Notebookcheck). NVIDIA’s CUDA is also the best-supported platform for AI tools. Good to know: it’s a heavy desk machine with a 400 W power brick, and 16–24GB caps the model size well below a 128GB Mac.

Check the Legion Pro 7i on Amazon →

3. ASUS ProArt PX13 GoPro Edition: big models on Windows

AMD’s Ryzen AI Max+ 395 works like Apple’s chips: the CPU and Radeon 8060S graphics share one pool of memory. In the ProArt PX13 GoPro Edition, that’s 128GB, and up to 96GB can go to the graphics, which is enough for much bigger models than any laptop RTX card. It’s a 3.1 lb, 13.3-inch convertible with a 3K OLED touchscreen (The Gadgeteer). Good to know: AMD’s software support for AI tools is less mature than NVIDIA’s CUDA, so check that your apps support it, and Tom’s Guide notes the screen is only 60Hz.

Check the ProArt PX13 GoPro Edition on Amazon →

4. Lenovo Legion Pro 5i Gen 10 (RTX 5070 Ti): best value

An RTX 5070 Ti’s 12GB of video memory is the sweet spot for 7–14B models, which cover a lot of everyday chat and coding use. Lenovo runs it at the full 140 W in the Legion Pro 5i, with a 16-inch 2.5K OLED screen, and in Notebookcheck’s testing the 5070 Ti version was up to 25% faster in games than the RTX 5070. See our best RTX 5070 Ti and 5080 laptops guide for more options. Good to know: avoid 8GB RTX 5070 models if local AI is the main reason you’re buying.

Check the Legion Pro 5i (RTX 5070 Ti) on Amazon →

5. MacBook Air 13 (M5) with 24 or 32GB: for small models

If you want to try small local models on an everyday laptop, a MacBook Air M5 with 24GB or 32GB of memory runs them silently and lasts all day. It starts with 16GB; configure more memory if AI is part of the plan, since it can’t be upgraded later (Apple). Good to know: it has no fan, so long generations slow it down, and the base M5’s 153GB/s of memory bandwidth makes it much slower than an M5 Max or RTX card. See our MacBook Air M5 page.

Check the MacBook Air (M5) on Amazon →

Tips before you buy

• Decide on the model size first, then buy the memory to fit it with room to spare for longer conversations.
• Check the graphics wattage on Windows. The same RTX chip runs much slower in thin laptops; our laptop GPU power database lists the wattage of 200+ configurations.
• Buy plenty of storage. Models are large files; 2TB is a sensible minimum if you’ll keep several.
• Don’t count on the NPU. The small AI accelerators in “Copilot+” laptops are built for low-power features, while most local LLM apps run on the GPU or CPU.

Our verdict

Buy for memory. The MacBook Pro with M5 Max and up to 128GB of unified memory runs the largest models of any laptop. On Windows, an RTX 5090 (24GB) or RTX 5080 (16GB) laptop such as the Legion Pro 7i is fastest for models that fit, the ASUS ProArt PX13’s 128GB of shared memory takes much bigger ones, and an RTX 5070 Ti (12GB) laptop is the value pick. A MacBook Air with 24–32GB is fine for experimenting with small models.

Frequently asked questions

Related guides

• Best RTX 5070 Ti and RTX 5080 laptops
• Laptop GPU power database
• Best laptops for programming
• Best MacBook for video editing
• Best RTX 5090 laptops

The best laptop for most programmers is the MacBook Air 13 (M5): 16GB of memory as standard, a Unix-based OS, a full day of battery and no fan noise. If you need Linux, the Framework Laptop 13 Pro is Framework’s first Ubuntu-certified laptop and lets you upgrade memory and storage later. For Windows and the best keyboard, get the Lenovo ThinkPad X1 Carbon Gen 14; for CUDA, machine learning or game development, you need an Nvidia GPU, such as the Lenovo Legion Pro 5i.

Check the Apple MacBook Air 13 (M5) on Amazon →

How we put this together: this guide is based on manufacturers’ specifications and published reviews (linked below). It is not based on hands-on testing by iTech Level. Prices are list or launch prices at the time of writing; memory costs have pushed many laptop prices up in 2026. See how we choose products.

Best laptops for programming compared

What matters in a programming laptop

• Memory first: we’d treat 16GB as the minimum. IDEs, browsers full of docs, Docker containers, virtual machines and Android emulators all add up, so choose 32GB if you run containers or VMs daily. On MacBooks memory can’t be upgraded later; on the Framework it can.
• The OS your work targets: macOS is Unix-based and covers web, mobile (including iOS, which needs a Mac) and most backend work. Linux suits server and systems work. Windows with WSL handles most things, and is needed for some .NET and Windows-only tools.
• A GPU only if you need CUDA: Nvidia’s CUDA, used by many machine learning tools, needs an Nvidia graphics card. Otherwise integrated graphics are fine.
• Keyboard, screen and battery: you’ll type and read all day, so a good keyboard, a sharp screen and a battery that lasts a workday matter more than peak benchmark numbers.

The best laptops for programming in 2026

1. Apple MacBook Air 13 (M5): best laptop for most programmers

The M5 MacBook Air now starts with 16GB of memory and 512GB of storage, which covers most development work, and it’s silent because it has no fan. Reviews are consistently positive: TechRadar calls it “still the best ultraportable I’ve ever used”, and RTINGS highlights its build, touchpad and battery life. Apple rates it at up to 18 hours of video streaming, and it drives two external displays. Students get education pricing. Our MacBook Air 13 (M5) guide has the full specs.

Good to know: with no fan it slows down under long, heavy loads such as big builds, and memory can’t be upgraded later, so buy 24GB or 32GB if you run containers. Skip the cheaper MacBook Neo for development: its 8GB of memory is fixed (see MacBook Neo vs MacBook Air).

Check the MacBook Air 13 (M5) on Amazon →

2. Apple MacBook Pro 14 (M5): best for long builds and heavy workloads

If you compile large projects, run test suites or keep several containers busy, the MacBook Pro’s fan lets the M5 hold its speed where the Air throttles. The base 14-inch model comes with 1TB of storage and adds a brighter XDR display, HDMI and an SD card slot. Memory goes up to 32GB on the M5; for more, step up to the M5 Pro models.

Good to know: it’s $700 more than the Air and heavier. Most web and app developers won’t feel the difference day to day.

Check the MacBook Pro 14 (M5) on Amazon →

3. Framework Laptop 13 Pro: best for Linux and upgradeability

Announced on April 21, 2026, the Framework Laptop 13 Pro is a ground-up redesign with a CNC-machined aluminum chassis, Intel Core Ultra Series 3 or AMD Ryzen AI 300 processors, and a 13.5-inch 3:2 2880×1920 touchscreen at 30–120Hz and up to 700 nits. It’s Framework’s first Ubuntu Certified system, with Ubuntu preinstalled as an option. Memory uses the new LPCAMM2 format, so you can upgrade from 16GB to 32GB or 64GB later, and storage goes up to 8TB. Framework claims over 20 hours of 4K streaming from its 74Wh battery, and it has four Thunderbolt 4 ports and weighs 1.4 kg (Framework).

Good to know: it started at $1,499 pre-built or $1,199 for the DIY Edition, but Framework has raised memory and storage prices during 2026 as memory costs soared (Framework), so check the current price. It’s sold mainly on Framework’s own site.

4. Lenovo ThinkPad X1 Carbon Gen 14: best Windows laptop for programmers

Developers who type all day care about the keyboard, and The Gadgeteer, which gave it an Editor’s Choice award, calls the X1 Carbon’s “the best typing experience in a business laptop right now”. The 2026 redesign weighs 2.15 pounds, runs Intel’s Panther Lake chip at up to 30 W thanks to a bigger fan, and scored 9 out of 10 for repairability at iFixit, with a customer-replaceable battery. There’s an optional 14-inch 2.8K 120Hz OLED.

Good to know: it’s expensive, especially in the Core Ultra 7 / 32GB / OLED configuration The Gadgeteer reviewed. The previous Gen 13 is usually cheaper. More business options are in our best business laptops.

Check the ThinkPad X1 Carbon on Amazon →

5. Lenovo ThinkPad E14 Gen 7 (Intel): best budget laptop for programming

For students and beginners, the ThinkPad E14 gives you the ThinkPad keyboard and TrackPoint, good maintenance access and modern Intel chips for much less. Notebookcheck says the Intel versions are the ones to buy: the Core Ultra 5 model with the upgraded 400-nit, 120Hz screen came to about €750 in Europe.

Good to know: avoid the AMD versions, which Notebookcheck found use older chips and a dim 300-nit screen. Make sure the configuration you buy has at least 16GB of memory.

Check the ThinkPad E14 Gen 7 on Amazon →

6. Lenovo Legion Pro 5i Gen 10: best for CUDA, machine learning and game development

If your work uses CUDA, for training or running machine learning models locally, or you build games in Unreal or Unity, you need an Nvidia GPU. The Legion Pro 5i runs its RTX 5070 at the full 115 W and its RTX 5070 Ti at 140 W, according to Lenovo’s spec sheet, next to an Intel Core Ultra 7 255HX and a 16-inch 2.5K OLED. The RTX 5070 Ti version has 12GB of video memory, which matters for larger models. See our best RTX 5070 laptops and GPU power database for alternatives.

Good to know: gaming laptops are heavy and their batteries don’t last long away from a socket. For serious model training, a desktop GPU or cloud instance will still be faster.

Check the Legion Pro 5i on Amazon →

Best Linux laptops for developers

Almost any laptop can run Linux, but the ones below are sold with it or certified by the maker, which means working Wi-Fi, graphics and sleep out of the box, plus support if something breaks. Dell, for example, only supports Ubuntu on systems that shipped with it. If you’d rather install it yourself, check Canonical’s Ubuntu certified laptops list first.

Linux options and prices from each maker’s site, checked October 23, 2026; preinstalled Linux is often US-only, so check your region’s configurator.

Our verdict

The MacBook Air 13 (M5) is the best laptop for most programmers. Choose the MacBook Pro 14 for long builds, the Framework Laptop 13 Pro for Linux and upgradeable memory, the ThinkPad X1 Carbon Gen 14 for Windows and the best keyboard, the ThinkPad E14 Gen 7 on a budget, and the Legion Pro 5i if you need CUDA. Whatever you pick, get at least 16GB of memory, and 32GB if you run containers or VMs.

Frequently asked questions

Related guides

• Best business laptops
• MacBook Air 13 (M5)
• Best laptops for video editing
• Best RTX 5070 laptops
• Best laptops for running local AI models

Short answer: buy the MacBook Neo if you mostly browse, write, stream and video-call: it has the same 2.7 lb weight and an almost identical screen to the Air, for $600 less. Buy the MacBook Air 13 (M5) if you edit photos or video, keep lots of apps open, use two external monitors, or want the laptop to last many years: in Tom’s Guide’s tests it was about twice as fast in multi-core work and lasted two hours longer on battery.

Buying for the office? Our best laptops for work guide compares MacBooks with ThinkPad, EliteBook and Dell Pro models.

How we put this together: this comparison uses Apple’s specifications, Tom’s Guide’s side-by-side lab tests and published reviews (linked below). It is not based on hands-on testing by iTech Level. Prices are Apple’s US prices after its June 25, 2026 price increase.

MacBook Neo vs MacBook Air: specs compared

Specs from Apple (MacBook Neo, launch details via MacRumors) and our MacBook Air 13 (M5) guide. Battery results from Tom’s Guide.

Price: the Neo saves you $600

Apple raised prices across its Macs on June 25, 2026, citing memory costs: the Neo went up by $100 and the Air by $200 (MacRumors). That leaves a $600 gap, with or without education pricing. The Neo’s 512GB model doubles the storage and adds Touch ID for $100 more, which still leaves it $500 below the Air.

Performance: the Air is about twice as fast under load

In Tom’s Guide’s tests, the Air scored 4,191 single-core and 17,276 multi-core in Geekbench 6, against 3,535 and 8,920 for the Neo. Exporting a video in Handbrake took the Air 4 minutes 34 seconds and the Neo nearly 10 minutes, and the Air’s SSD was over four times faster (about 6,500 MB/s against 1,500 MB/s).

What that means in practice: for web browsing, Google Docs or Office, Zoom and streaming, the single-core gap is small and the Neo feels quick. The difference shows when you export video, edit large photo libraries, run many apps at once or compile code, and the Neo’s 8GB of memory can’t be upgraded later.

Display: nearly a tie

Tom’s Guide measured both screens at about 452–454 nits in SDR. The Air covered slightly more color (83% of DCI-P3 against 78.6%) and its 13.6-inch panel is a little larger and sharper, but for everyday use you won’t see much difference. The bigger display gap is external: the Neo drives one monitor up to 4K 60Hz, through its USB 3 port only, while the Air drives two.

What the Neo gives up

• No keyboard backlight, which makes typing in the dark harder unless you touch-type.
• No MagSafe: you charge over USB-C, which takes one of its two ports, and the second port runs at slow USB 2 speed.
• No Touch ID on the base model; you need the 512GB version.
• A simpler camera: 1080p, against the Air’s 12MP Center Stage camera.
• 8GB of memory, fixed for life, and a slower SSD.

Battery life: the Air lasts longer

Apple rates the Neo at 16 hours and the Air at up to 18 hours of video streaming. In Tom’s Guide’s web-surfing test at 150 nits, the Neo lasted 13 hours 28 minutes and the Air 15 hours 37 minutes. Both last a full school or work day; the Air gives you more margin.

Which should you buy?

Buy the MacBook Neo if…

…you’re a student, a first-time Mac buyer, or buying for a child, and you mainly browse, write, stream and video-call. Tom’s Guide recommends it for exactly those people. It’s the cheapest way into macOS, and with education pricing it competes with good Chromebooks; see our best Chromebooks if you’re weighing the two.

Check the MacBook Neo on Amazon →

Buy the MacBook Air 13 (M5) if…

…you edit photos or video, work with many apps and browser tabs open, connect two monitors, want MagSafe and a backlit keyboard, or plan to keep the laptop for five years or more. Twice the memory and storage, a much faster chip and longer battery life justify the extra $600 for that kind of use. If you edit video regularly, also see our best MacBook for video editing.

Check the MacBook Air 13 (M5) on Amazon →

Black Friday: what to expect

Black Friday 2026 is November 27. Discounted M4 Airs will be common, but compare like for like: the base M4 Air has only 256GB of storage. Our Black Friday laptops guide covers what’s worth waiting for.

Frequently asked questions

Related guides

• MacBook Air 13 (M5): specs and reviews
• Best MacBook for video editing
• Best Chromebooks
• Best business laptops and laptops for work
• Best laptops to buy on Black Friday