I recently stumbled upon a post that mentioned GPT-OSS:20B, and I was curious to learn more about it. After some research, I discovered Ollama, a tool that allows you to run GPT-OSS:20B locally. I was intrigued by the idea of having a powerful AI model at my fingertips, so I decided to give it a try.
I downloaded and installed Ollama on my desktop, which has a Ryzen 7 processor, 32GB of RAM, and an old GTX 1080 GPU. While it’s not the most powerful machine, I was surprised to find that Ollama works reasonably well, albeit a bit slowly.
What I like about Ollama is the idea of running it locally. It gives me more control over my data and allows me to experiment with AI models without relying on cloud services. But I do have some questions – is Ollama truly running locally, or is it still dependent on cloud infrastructure?
I’d love to hear from others who have experience with Ollama. What are your thoughts on running AI models locally, and do you think it’s the future of AI development?