There are some factors to consider. Some of the Deepseek quants are based on Llama 3, whereas others are based on Qwen Reasoning.

You’re also not going to get the same quality of the full ChatGPT experience comparing a 7B parameter model to a 500B+ model like ChatGPT.

Regardless, it’s difficult to run the actual Deepseek R1 model as there’s not a true quantization or distillation of the original model.

You can also try GPT-OSS if you want an open source model comparable to ChatGPT. Once again, you’re going to have to balance the size and precision of the model with your expectations.

Seconded - just use Tailscale and SSH.

Ansible!

I second GitLab CI/CD - it’s a CI/CD system that just makes sense to me. That doesn’t mean it doesn’t have its complexities depending on your needs, but I’ve overall enjoyed my time working with it.

Sorry for spamming in this thread, but if you rely on Watchtower, there’s a maintained fork I recommend: https://github.com/nicholas-fedor/watchtower

Not for the latest and future versions of Docker.

This fork works, though: https://github.com/nicholas-fedor/watchtower

There’s a maintained fork, fortunately!

https://github.com/nicholas-fedor/watchtower

Deepseek R1 and OpenThinker are two more examples. There’s also SmolLM, which I believe also open sources its training data and ensures proper licensing for it.

I have a similar perspective. I built my own in-home AI server because I assumed if the technology had any staying power, I better learn how it works to some degree and see if I can run it myself.

Jan is another great recommendation!

I’m keeping an eye on Ollama’s service offerings - I don’t think they’re in enshittification territory yet, but I definitely share the concern.

I still don’t believe the other LLM engines out there have reached an equivalent ease of use compared to Ollama, and I still recommend it for now. If nothing else, it can be a stepping stone to other solutions for some.

In case you’re not aware, there are a decent number of open weight (and some open source) large language models.

The Ollama project makes it very approachable to download and use these models.

Exactly!

There’s not a significant amount of discourse suggesting that this is a natural cycle: https://news.cornell.edu/stories/2021/10/more-999-studies-agree-humans-caused-climate-change

Rhetoric like yours adds uncertainty where there is extremely little to be had.

These cycles also perpetrate over millennia, not decades, which is the current scope of detectable change we’re dealing with.

1. Install Ollama
2. ollama pull hf.co/mradermacher/VibeThinker-1.5B-GGUF
3. ollama run hf.co/mradermacher/VibeThinker-1.5B-GGUF

If this has a gyroscope, I’m sold.

The login page for the old apps is at the bottom of the page in the footer.

That being said, I really miss this being a premium offering.

“Free”

Try using it standing up.

I think cooler is subjective. With a physical keyboard back in the day and Remote Desktop, I had a pocket-sized Windows PC with me at all times. With SSH, I had a portable terminal I could easily administer servers around the world with. I thought that was pretty cool.

Now I’m tap typing on a device with no physical feedback where the keyboard hides half the screen and reshuffles my terminal output every time said keyboard is shown and hidden. That’s not cool at all.