I mean, maybe that is a possibility. I just never thought of it. I would recommend checking the official documentation. Might help shed some light on this.

I don’t think so, no.

Awesome

I am not entirely sure. My best piece of advice here would be: check their in depth documentation on GitLab. It might answer your questions.

That’s great news! I’d love for it to be added to a wiki. Just make sure that whatever version of this post is added to the wiki is the most updated one.

No. That has to do with how the Tor network works. The bridge forwards the connection to a non exit relay. You do not communicate with an exit relay whatsoever. The middle relay does, but the exit relay doesn’t know who are are and you don’t know who the exist relay is.

You are completely right. That was worded poorly and a few users have thankfully pointed that out. The answer, for most people, is yes. But that depends entirely on your threat model.

The traffic to your Snowflake proxy isn’t necessarily from people in ‘adversarial countries’. A Snowflake proxy is a type of bridge, so just about anyone can use it. You can use a Snowflake bridge, if you want. However, it is strongly encouraged to save bridges (including Snowflakes) to people who need them.

So, for most people, it is generally safe to run Snowflake proxies. Theoretically, your ISP will be able to know that there are connections being made there, but, to them, it will look like you’re calling someone on, say, Zoom since it uses WebRTC technology. They can’t see the data, though since everything is encrypted (check the Snowflake docs and Tor Brower’s for further reference). You probably won’t get in any trouble for that.

Historically, as far as we know, there haven’t been any cases of people getting in legal trouble for running entry relays, middle relays, or bridges. There have a been a few cases of people running exit nodes and getting in trouble with law enforcement, but none of them have been arrested or prosecuted so far.

If you know of any, let me know.

I have not used AI to write the post. I used Claude to refine it because English is not my first language. If there are any errors, that is my bad. Please point them out as you did so I can fix them.

This has several errors including the fact that running the proxy exposes your IP address.

Thank you for pointing that out. That was worded pretty badly. I corrected it in the post.

For further clarification:

The person who is connecting to your Snowflake bridge is connecting to it in a p2p like connection. So, the person does know what your IP address is, and your ISP also knows that the person’s IP address is – the one that is connecting to your bridge.

However, to both of your ISPs, it will look like both of you are using some kind of video conferencing software, such as Zoom due to Snowflake using WebRTC technology, making your traffic inconspicuous and obfuscating to both of your ISPs what’s actually going on.

To most people, that is not something of concern. But, ultimately, that comes down to your threat model. Historically, there haven’t any cases of people running bridges or entry and middle relays and getting in trouble with law enforcement.

So, will you get in any trouble for running a Snowflake bridge? The answer is quite probably no.

For clarification, you’re not acting as an exit node if you’re running a snowflake proxy. Please, check Tor’s documentation and Snowflake’s documentation.

deleted by creator

Though apparently I didn’t need step 6 as it started running after I downloaded it

Hahahha. It really is a little redundant, now that you mention it. I’ll remove it from the post. Thank you!

Good fun. Got me interested in running local LLM for the first time.

I’m very happy to hear my post motivated you to run an LLM locally for the first time! Did you manage to run any other models? How was your experience? Let us know!

What type of performance increase should I expect when I spin this up on my 3070 ti?

That really depends on the model, to be completely honest. Make sure to check the model requirements. For llama3.2:2b you can expect a significant performance increase, at least.

I have tried on more or less 5 spare phones. None of them have less than 4 GB of RAM, however.

Great explanation, Max!

I would argue there would not be any noticeable differences.

The performance may feel somewhat limited, but this is due to Android devices usually having less processing power compared to computers. However, for smaller models like the ones I mentioned, you likely won’t notice much of a difference when running them on a computer.

Pretty much. It’s pretty straight forward.

Hmmm… You’re right. It does feel a lot more arbitrary when you put it that way.

My favorite anime website is down; good thing FMHY has a bunch of great ones to choose from. Migrating sucks, though.

There isn’t really a natural barrier between North and South America, though. Asia has the Urals.

Interesting question… I think it would be possible, yes. Poison the data, in a way.

Not Perplexity specifically; I’m taking about the broader “issue” of data-mining and it’s implications :)