Installing Llama on Ubuntu 26.04 and 24.04: Commands, Models, and RAM

On Ubuntu, the shortest documented route to running a Llama model locally is to install Ollama and run a model from its library. Ollama provides a Linux install command and a Llama 3.2 command; Ubuntu also documents installing Ollama as a Snap. The commands below are documented guidance, not a compatibility test for every Ubuntu 26.04 or 24.04 system.

What to install on Ubuntu

Ubuntu 24.04 LTS and 26.04 LTS are both listed in Ubuntu’s documentation index. A fresh Ubuntu 26.04 Desktop installation does not include built-in AI tooling, according to the Ubuntu AI page, last edited September 25, 2026. Installing a runtime such as Ollama is therefore a separate step.

#	Preview	Product	Price
1		GEEKOM Air12 Budget Mini PC Office,Intel 7505,8GB RAM(64GB Max),256GB SSD	$499.00	Buy on Amazon

As an Amazon Associate I earn from qualifying purchases.

Ollama is the primary path here because its Linux quickstart documents both installation and a Llama run command. Its instructions are Linux guidance; the reviewed documentation does not provide a compatibility matrix or separate test results for Ubuntu 24.04 and 26.04.

Install Ollama and run Llama

Linux install-script route

Ollama documents this command for Linux installation:

#1 Best Overall

GEEKOM Air12 Budget Mini PC Office,Intel 7505,8GB RAM(64GB Max),256GB SSD

• ➊ [ Trusted Quality for Everyday Agentic AI ] GEEKOM equips its SSDs with reliable original-grade flash and conducts rigorous stability testing to support dependable everyday operation. This commitment to quality is backed by a 3-year warranty. Simply connect the Air12 to cloud AI services for research, writing, study support and daily productivity—no NPU or complex local setup required. Designed for students, home users, light office work and first-time buyers, the Air12 is a high-value Cloud Agentic PC for everyday tasks
• ➋ [ Intel 7505 processor ] Powered by the Intel 7505 processor (2 cores, 4 threads, up to 3.5GHz), the GEEKOM Mini PC Air12 delivers smooth performance for everyday computing, office tasks, and home entertainment. With enhanced single-core processing, it handles daily workloads efficiently and responsively. Compact, quiet, and energy-efficient — a solid alternative to bulky desktops.
• ➌ [440lbs(200kg) Pressure Rated Metal Frame for Demanding Environments] Unlike the Plastic Shells You’ll Find on Most Mini PCs, geekom Mini Air12 features a triple-reinforced ABS+PC shell, precision-crafted metal frame and baseplate—engineered to withstand up to 440 lbs of pressure for the perfect balance of strength and thermal efficiency. Tool-free upgrades, shock-absorbing feet, and a 3D antenna deliver true durability
• ➍ [Dual-Channel RAM & NVMe SSD Expandability] Ships with 8GB DDR4 RAM and a 256GB NVMe SSD for smooth everyday performance. Dual memory slots and dual storage slots give you the flexibility to upgrade to 64GB RAM and 2TB SSD, so your system can adapt as your workload grows. Enjoy faster load times, smoother multitasking, and long-term reliability.
• ➎ [Triple 4K Displays for Maximum Productivity] Connect up to three 4K monitors via HDMI 2.0, Mini DisplayPort 1.4, and USB-C — ideal for stock trading dashboards, multi-tab research, office document editing, and light spreadsheet work. WiFi 6 and Bluetooth with high-gain antenna ensure stable wireless connections throughout your workspace. 5x USB ports and a full-size SD card reader provide quick access to peripherals and camera files — no adapters required.

$499.00

Buy on Amazon

curl -fsSL https://ollama.com/install.sh | sh

When installation is complete, run Llama 3.2:

ollama run llama3.2

Ollama says this starts an interactive chat and downloads the model if it is not already available locally. These are commands from its documentation, not commands independently verified on each Ubuntu release. See the Ollama Linux instructions and Ollama quickstart.

Ubuntu Snap route

Ubuntu’s AI-on-Ubuntu page documents installing Ollama from Snap:

sudo snap install ollama

Ubuntu’s page uses Qwen in its sample run command, not Llama. To use Llama after installing the Snap, use Ollama’s documented model command:

ollama run llama3.2

This combines Ubuntu’s documented Snap installation route with Ollama’s Llama command; the Ubuntu example itself is not a Llama walkthrough. See Ubuntu’s AI on Ubuntu page.

Useful Ollama commands

Ollama documents these commands for managing a local model:

• ollama pull llama3.2 downloads or updates the model; Ollama says only the difference is downloaded when updating.
• ollama list lists locally available models.
• ollama show llama3.2 displays model information.
• ollama ps shows running models.
• ollama stop llama3.2 stops the running model.
• ollama rm llama3.2 removes the local model.

Ollama also documents ollama serve for starting the service without the desktop application, and a local REST API at http://localhost:11434, including /api/generate and /api/chat endpoints. Consult the Ollama quickstart for the documented operations.

Llama model examples and listed download sizes

Ollama’s quickstart lists the following model examples, commands, parameter counts, and download sizes. The GB values are the page’s listed download sizes; they are not total installed disk use, RAM requirements, GPU memory requirements, or speed estimates.

These listings come from Ollama’s rolling quickstart documentation, accessed in 2026; model library tags and displayed sizes can change. Check the Ollama model library and quickstart if the exact current listing matters.

How much RAM do Llama models need?

Ollama gives broad available-RAM guidance by parameter count: at least 8 GB for 7B models, 16 GB for 13B models, and 32 GB for 33B models. This is guidance from Ollama’s quickstart, not a guarantee of fit, speed, or a GPU requirement. It also does not establish exact minimum RAM for every Llama variant in the table above.

Parameter count, downloaded model size, installed storage use, system RAM, and GPU VRAM are different measures. The download figures above cannot be used as a direct substitute for memory requirements. Context length and hardware configuration can also affect whether a particular setup is suitable; the cited guidance does not specify a context length or device configuration.

Using llama.cpp instead

For readers who want a command-line workflow based on model files, llama.cpp supports models in GGUF format. Its model documentation describes fetching a compatible model with -hf <user>/<model>[:quant] or running a local GGUF file. Models in other formats need conversion using project scripts.

The project’s documented example is for Gemma, not Llama:

llama cli -hf ggml-org/gemma-3-1b-it-GGUF

That example should not be treated as a Llama command. The documentation establishes the GGUF workflow but does not provide a verified current Llama repository/tag and end-to-end command here. Quantized GGUF sizes depend on the model and quantization, so Ollama’s listed download sizes do not apply. See the llama.cpp model documentation for its workflow and links to quantization information.

Choosing a route and setting expectations

The available documentation supports comparing installation and model-file workflows, but not ranking these options for speed, output quality, GPU use, or privacy guarantees. Hardware acceleration can depend on the computer and its drivers; no GPU-specific recommendation or Ubuntu 26.04/24.04 acceleration result is established by these sources.

Quick Recap

Bestseller No. 1

GEEKOM Air12 Budget Mini PC Office,Intel 7505,8GB RAM(64GB Max),256GB SSD

$499.00

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.