Comply with ZDNET: Add us as a preferred source on Google.
ZDNET’s key takeaways
- Ollama is a user-friendly, domestically put in AI.
- You’ll be stunned at the advantages you acquire from this app.
- Not solely is Ollama free and personal, nevertheless it’s additionally open-source.
Ollama is a small participant within the AI sport, nevertheless it needs to be a lot larger than it’s. This installable AI has a number of advantages you will not discover with the likes of ChatGPTand people advantages are what maintain me from ever utilizing a extra conventional service.
I’ve written about Ollama very often over the previous 12 months, however I assumed it was time to share the explanations I select to make use of this app over another.
Additionally: Want local vibe coding? This AI stack might replace Claude Code and Codex – for free
(Disclosure: Ziff Davis, ZDNET’s guardian firm, filed an April 2025 lawsuit in opposition to OpenAI, alleging it infringed Ziff Davis copyrights in coaching and working its AI programs.)
What’s Ollama?
Earlier than I get into the why, let’s speak concerning the what.
Ollama is a free, open-source software which you can obtain and set up in your pc (Linux, MacOSor Windows). With Ollama, you possibly can run massive language fashions (LLMs) instantly in your system. The principle challenge to think about is your pc’s pace. AI requires a good bit of horsepower to be helpful. That does not imply you possibly can’t apply it to a midrange machine, however should you’re utilizing it on mid- to low-end {hardware}, it’s going to run extra slowly, and also you may not be capable of multitask whereas it churns via your queries.
Additionally: OpenAI’s new image watermarks make it easier to spot AI fakes – here’s how
Nonetheless, in case your machine has an Nvidia GPU, it is going to run sooner (even with midrange {hardware}) as a result of it’s going to offload loads of the processes onto the GPU, leaving the CPU to do different issues. To run Ollama easily, your system ought to have a minimal of:
- CPU: Any trendy processor
- System RAM: 16 GB
- GPU (Really helpful): Nvidia with 8GB+ VRAM or an Apple Silicon Mac (M1/M2/M3) with 16GB+ unified reminiscence for computerized GPU acceleration
Ollama presents a user-friendly GUI for each MacOS and Home windows, however can be used from the command line. On Linux, you need to use the Ollama CLI, however there are additionally GUIs (reminiscent of Alpaca and Revenge) that may work.
With Ollama, you possibly can obtain many various LLMs. There is a huge library of models that can be downloaded — you will discover DeepSeekGemma, Qwen, Mistral, Gpt-OSS, Llama, and lots of others.
Now that you just perceive what Ollama is, let’s discuss why you ought to be utilizing it.
1. It is free
Ollama is free. Since I have been utilizing it, I’ve not paid a single penny for the app or the fashions. Simply set up Ollama, pull a mannequin, and go. You do not have to pay for the app, the fashions, or utilization.
It is free. Completely and eternally free.
It is also open-source, so… bonus.
Additionally: Google’s AI features just got more confusing
2. It is non-public
This can be a huge one for me. I will not use public or for-profit AI as a result of I do know my queries and the responses can be utilized by third events. The businesses behind for-profit AI can acquire your data and use it for varied functions, and that is not one thing I’m OK with. It isn’t that I question with any private data (I principally use AI for analysis), however I nonetheless don’t desire these queries collected and used to create a profile of me.
Additionally: The best secure browsers for privacy: Expert tested
Identical to my browsersI would like my AI to be as non-public as attainable, and Ollama is basically the one approach to make that occur.
3. It would not problem {the electrical} grid
You have heard of how AI makes use of an inordinate quantity of power, proper? A report this month from the International Data Center Authority (ICDA) notes that an estimated 6% of whole US electrical energy use goes to knowledge facilities. “The US is by far the world’s largest knowledge heart location, with 43% of world consumption. Knowledge facilities eat 29.2GW of electrical energy within the US, and now eat 6 % of the nation’s electrical energy.”
This A postponement particulars the environmental issues surrounding large-scale AI deployments. “Most AI servers are saved in knowledge facilities, which produce digital waste and may include poisonous chemical compounds, reminiscent of mercury and lead. Knowledge facilities additionally require massive quantities of water for development and to chill {the electrical} elements. “
Additionally: Worried about AI’s soaring energy needs? Avoiding chatbots won’t help – but 3 things could
By utilizing locally-installed Ollama, I keep away from including to that downside. I might take that one step additional and set up Ollama on a laptop computer, so I am solely utilizing battery energy.
Ollama is environmentally pleasant and doesn’t put any pressure on {the electrical} grid.
4. It is extra versatile
While you’re utilizing the likes of ChatGPT, you are restricted to the LLMs you possibly can select from. With Ollama, you’ve gotten a large library of LLMs at your fingertips. You’ll be able to even set up a number of fashions and select which one to make use of for every question.
Additionally: How I easily added AI to my favorite Microsoft Office alternative
That is a flexibility you will not discover in most for-profit AI companies.
5. It is LAN-able
I’ve one occasion of Ollama put in on a server inside my LAN. With that up and operating, I can both use a web-based interface or join a neighborhood GUI app to the server, so I am not utilizing CPU/GPU on my laptop computer or desktop. That is an awesome possibility as a result of offloading the AI course of to a server means I haven’t got to fret about queries bringing much less highly effective {hardware} to a halt. On high of that, I solely have to put in Ollama on one machine to apply it to each machine linked to my LAN.
6. Offline accessible
To illustrate you lose your web connectivity, otherwise you’re out in the midst of nowhere with out a sign or Wi-Fi, and also you want (for no matter purpose) to make use of AI. When you use a third-party, cloud-based AI, you are out of luck. When you’ve got Ollama put in in your laptop computer, you are very a lot in luck, as a result of you do not have to have a community connection to make use of it. That additionally means you possibly can run Ollama on an air-gapped machine for much more privateness.
Additionally: How I feed my files to a local AI for better, more relevant responses
It is win-win throughout.
I might extremely advocate you strive a domestically put in occasion of Ollama. It is a lot simpler to stand up and operating than you suppose (just like putting in any app on MacOS or Home windows), it is low cost, versatile, safe, and environmentally pleasant.
Who would not need all of that?
