A “QuitGPT” campaign is urging people to cancel their ChatGPT subscriptions— Backlash against ICE is fueling a broader movement against AI companies’ ties to President Trump.
Having just finished getting an entire front end for my website, I disagree. A few years ago I would offshore this job to some third-world country devs. Now, AI can do the same thing, for cents, without having to wait for a few days for the initial results and another day or two for each revision needed
Please enlighten me. I am working on systems solving real-world issues, and now I can ship my solutions faster, with lower costs. Sounds like a win-win for everyone involved except for the offshore employees that have to look for new gigs now
Edit: I would actually rather read a reply than just see you downvoting. The point is, what you call a “glorified text generating machine”, has actual use cases
I don’t know if it’s your fault honestly. It’s the system that makes you want to offshore your work to developing countries and not hire local employees. I get it. It’s cheaper. But when even independent developers start doing this we have reached post-late stage capitalism at this point
it's not that it doesn't have use cases. It burning the world down with it.
how much more water went down the drain cooling the requests you made? how about the electricity not going to local consumers but AI data centers.
all the computer components shortages...
that's still before the fact you admitted you would have hired a human, and given them food on the table instead of a corporate giant to buy another mega yacht.
Regarding the electricity not going to local customers: it’s not my fault that your country does not have appropriate regulations. None of the data centers are located where I live anyways.
Regarding hiring a human: I 100% would hire a human if there was no AI, you’re right, I’m not trying to hide it. Sucks for them I guess but I don’t see a reason why should I keep using their services if I can get a cheaper and arguably better alternative now - I’m trying to make money, not run a charity supporting development of 3rd world countries with authoritarian regimes.
However, I am pretty sure that the companies which I’m paying are currently operating on a loss with my $3/month and free coding plans. They want to grow the customer base and I’ll just switch once they will start wanting to make a profit, like the Chinese ZAI just did this week, hiking up their prices by over 3x. They are operating on a loss with my <3$/month contribution and not buying another mega yacht yet getting further away from one.
Looks like others have come along and made my point for me while I slept. Except for calling out the dehumanizing language against the developers. They missed that one.
They weren’t my full time employees but one time gigs. I’ve worked with tons of freelancers over the years, I don’t have any special relationship with them. It’s was just a matter of whoever would offer the lowest price for the gig each time
Yeah. It's not hard to say a sentence about the problem space. Heck, in good faith I'll say what I've been working on to start. I'm currently developing tools to help small communities teach their native language to younger generations as existing programs have stopped support for them.
It’s behind a login page so I’m afraid you wouldn’t see much. Also, it was never supposed to be glorious (it was not before the LLMs neither), it’s a matter of just having some form of UI as a necessity. I would be hiring actual designers if it was supposed to be a landing page or sth where the looks matter, not stick with the AI
So yours and another comment I saw today got me to dust off an old docker container I was playing with a few months ago to run deepseek-r1:8b on my server's Intel A750 GPU with 8gb of VRAM. Not exactly top-of-the-line, but not bad.
I knew it would be slow and not as good as ChatGPT or whatever which I guess I can live with. I did ask it to write some example Rust code today which I hadn't even thought to try and it worked.
But I also asked it to describe the characters in a popular TV show, and it got a ton of details wrong.
8b is the highest number of parameters I can run on my card. How do you propose someone in my situation run an LLM locally? Can you suggest some better models?
To be clear, I think getting an LLM to run locally at all is super cool, but saying "go self hosted" sort of gloms over the fact that getting a local LLM to do anything close to what ChatGPT can do is a very expensive hobby.
I agree, it is a very expensive hobby, and it gets decent in the range 30-80b. However, the model you are using should not perform that bad, it seems that you might be hitting a config issue. Would you mind sharing the cli command you use to run it?
So I'm not using a CLI. I've got the intelanalytics/ipex-llm-inference-cpp-xpu image running and hosting LLMs to be used by a separate open-webui container. I originally set it up with Deepseek-R1:latest per the tutorial to get the results above. This was straight out of the box with no tweaks.
The interface offers some controls settings (below screenshot). Is that what you're talking about?
Those values are most of what I was looking for. An LLM is just predicting the next token (for simplicity, a word). It does so by generating every possible word with a probability associated with it, and then picking a random word from this list, influenced by its probability. So for the sentence “a cat sat” it might generate “on: 0.6”, “down: 0.2”, and so on. 0.6 just means 60%, and all the values add up to 1 (100%). Now, the number of tokens generated can be as big as the context, so you might want to pick randomly from the top 10, you control this with the parameter top_k, or you might want to discard all the words below 20%, you control this with min_p. And finally, in cases where you have a token with a big probability followed by tokens with very low probability, you might want to squash these probabilities to be closer together, by decreasing the higher tokens and increasing the lower tokens. You control this with the temperature parameter where 0.1 is very little squashing, and 1 a lot of it. In layman terms this is the amount of creativity of your model. 0 is none, 1 is a lot, 2 is mentally insane.
Now, without knowing your hardware or why you need docker, it is hard to suggest a tool to run LLMs. I am not familiar with what you are using, but it seems to not be maintained, and likely lacks the features needed for a modern LLM to work properly.
For consumer grade hardware and personal use, the best tool these days is llamacpp, usually through a newbie friendly wrapper like LMStudio which support other backends as well and provide so much more than just a UI to download and run models. My advice is to download it and start there (it will download the right backend for you, so no need to install anything else manually).
I'm running it in docker because it's running on a headless server with a boatload of other services. Ideally whatever I use will be accessible over the network.
I think at the time I started, not everything supported Intel cards, but it looks like llama-cli has support form Intel GPUs. I'll give it a shot. Thanks!
It goes down to number of vram / unified ram you have. There is no magic to make 8b perform like top tier subscription based LLMs (likely in 500b+ range, wouldn't be surprised if trillions).
If you can get to 32b / 80b models, that's where magic starts to happen.
Honestly you pretty much don't. Llama are insanely expensive to run as most of the model improvements will come from simply growing the model. It's not realistic to run LLMs locally and compete with the hosted ones, it pretty much requires the economics of scale. Even if you invest in a 5090 you're going to be behind the purpose made GPUs with 80GB VRAM.
Maybe it could work for some use cases but I rather just don't use AI.
Going local is taxing on your hardware that is extremely expensive to replace. Hell, it could soon become almost impossible to replace. I genuinely don‘t recommend it.
Even if you HAVE to use LLMs for some reason, there are free alternatives right now that let Silicon Valley bleed money and they‘re quickly running out of it.
Cancelling any paid subscription probably hurts them more than anything else.
Last I checked it costed ~$6000 to run a high end LLM at terrible speeds and that was before the RAM price hike. And at the rate things are changing it might be obsolete in a few year. And I don't have that much money either.
I'm going to stick with the free OpenCode Zen models for now and maybe switch to OpenCode Black or Synthetic or whatever when they stop being available and use the open weight models there for now.
I would, if I found even a remotely good usecase for LLMs. Would be useful for contextual search on a bunch of API documentation and books on algorithms, but I don't want a sychophantic "copilot" or "assistant", that does job so bad I would be fired for, all while being called ableist slurs and getting blacklisted from the industry.
I only have a need for RAG and it works well, wouldn't say I need w copilot either. Everyone probably has some use case for this stuff but it's gonna be different for everyone, the basic llm conversation model it's really just a stepped up interactive Google
Stop using Amazon, Meta products, Netflix, Spotify, etc.
All of these massive corporations are actively making life worse for people, and it will only get worse and worse as people continue to stay subscribed.
The only option is to log off and find an alternative.
QuitGPT is in the mold of Galloway’s own recently launched campaign, Resist and Unsubscribe.
Resist and Unsubscribe encourages you to get away from Google and Amazon, as well as OpenAI. QuitGPT endorses both (Google directly, Amazon indirectly).
Resist and Unsubscribe is a holistic project; QuitGPT cuts out everything except for one product.
Resist and Unsubscribe doesn't inadvertently promote a single product; QuitGPT practically functions as a sneaky advertising campaign (kind of like how Larry David said he wouldn't invest in FTX).
It seems pretty clear which project is better, even if I don't agree with everything the guy behind it said:
Galloway argued that the best way to stop ICE was to persuade people to cancel their ChatGPT subscriptions.
Well the fucked up part is businesses also want to see how well their "investment" AI is doing, which means tracking how much employees use the AI and other cooked up metrics. You can't get away from it no matter how irritating it makes your job and life.
All the time at work. Free tier and paid. Translate shit from foreign customers. Extract text from screen shots. There’s lots of uses, just like any other tool. But it’s a tool, it’s not fucking magic or the devil. Just a tool.
Unless you're (not you PhoenixDog specifically, the general You referring to the reader) blocking your browser fingerprinting, using Ad blocker, DNS filtering from trusted sources, not using social media app, not installing apps that require Google Play Services and not using your credit card for digital payments then simply changing the IP address that the server sees doesn't do very much.
VPNs can be part of a solution but if you are not doing all of the other things then they do not hide who you are because of the plethora of other avenues that you can be tied to an existing advertising identity.
Yeah, VPN's are unfortunately essential. If I hadn't made it a project to request deletion of my info off the top 30 background check sites, I'd be using something like DeleteMe too.
Educational institutions are buying microsoft services as a trend and we need them to boycott also.
Getting bus pass in my university takes 5 business days so they can "save cost on students who don't use it."
Meanwhile everyone's student gmail implicitly has a subscription to copilot, gemini, and openai at the same time; it doesn't matter if you are using them or not.
Grok/Mechahitler owner likely has donated more to GOP/Trump rule. Anthropic CEO wrote an essay that outlined dangers of AI recently, but also underlined how the pure good US must develop Skynet to oppress all of the bad evil opposition to the US empire.
The entire industry is lining up to get US government military/surveillance applications to use the absurd level of datacenters/energy that are being committed to, because "Skynet now or China wins" is even more aggressive consensus than "too big to fail".
Yes agree, i have quit MSFT fully (switched to Ubuntu), Amazon prime is gone and no longer buying from them. Cancelled my Claude subscription, my gmail now auto forwards to my local server, switched my web hosting from AWS to Hetzner, Cancelled my netflix, back to library and jellyfin local server. Same with Spotify. No more Safeway or QFC - Costco is good enough and has the lowest prices locally, or occasionally Trader Joes. Its saving me a whole shit ton of money.
I don't pay for AI. There is no real value in it at massive scale for it to be paid for me. Plus it is harmful. I can local host AI for small tasks like cleaning up some data.
Yep! I feel way, way safer when I use something running on my laptop as opposed to feeding things into the bottomless hole that is Microsoft or Anthropic.
All I want is something to double-check the verbiage of an email to make sure I'm not coming off like an asshole.
Also, why paying for AI? You can use it for free, and when you finish your free daily limit, you just switch to another AI: Gemini, Perplexity, Claude, Le Chat, DeepSeek... rinse and repeat.
Honestly, people are probably going to start using it much less once they put in ads on the the low and free tiers.
They're realizing they can't turn profit over expenses when the VC money dries up without jacking price.
Same shit different fad. Offshoring was all the rage until the demand meant they started charging more, the quality went down, and the reality of cross-timezone management became obvious. Cloud was all the rage until the prices went up from demand and also kept running into outages, and now some companies are "re-inhousing" their infrastructure.
Cant' wait till they realize they also can't deliver at regular clip with only half the workers. Wont' be soon enough, alas.
Mind if I ask how it's helped you along? My lil bro used it to do basically all of his college work LMFAO. He also uses ai screening tools to make sure he changes shit enough that the professor doesn't catch it, well one day he screened a assignment from the professor and it was flagged as AI produced LMFAO
I have memory issues (it’s a brain issue) and struggle to do some basic things, but higher order thinking is natural for me. So ChatGPT does the things I struggle with, I review it and then use that information for my bigger goals. I got a lot done last year because of chatgpt, and rn I’m on the last leg of the project. I spent a few years trying to complete the project alone and didn’t get far because of the memory issues.
How ai is used is dependent on the user itself. Some people just don’t want to put effort like your brother, some people have disabilities or other issues that prevent them from doing bigger things they’re otherwise capable of doing.
I bought about a dozen epub comics. They were formatted with a hardcoded 600x450 width or so, maybe expecting a particular device. Having recently worked with epubs to format my own (word) book, I knew the format, and basically wanted to use Python standard library tools to unzip them, rip out some useless sizing/styling code (from hundreds of XHTML files), and zip them back up.
I hadn't used Python professionally in a few years, so this was an annoying back and forth to work out the process and remind myself of syntax, especially considering this was something I was just doing for a few of my own books. Instead, taking every important piece of this puzzle/process I'd researched, I instead described the problem to ChatGPT, specifically pointing it to the Python standard libraries I wanted to use. It gave me a one-page program that was mostly complete and I only needed to change in a few areas.
I don't think I'd ever pay money to AIs for a variety of reasons. I take that assistance as it comes, and could live without it.
I personally use AI to help me with my work. I used to work in an IT repair facility where I would use an LLM to help me troubleshoot problems. For example if a computer won't turn on and the motherboard is flashing and beeping, asking an LLM is 10x faster than having to use the internet to hunt for a user manual of one specific model of computer from 40 years ago, made by a brand which no longer exists.
Now I work in Computer Games Development after getting my degree. AI is extremely useful for helping me come up with ideas, making prototypes, and even generating small amounts of code in C++ and C#, which is what I mainly use. Games Dev isn't difficult. I can do it myself. But it can be very tedious. That's why AI is so helpful.
As for what your little brother does, i've experienced that. People generating code and having no idea what it does. It might help you pass your subjects in school, but it's only gonna screw you over in the long run.
I don't mind people using AI to help them with their work as long as they don't depend on it.
Probably because it challenges the narrative of people who automatically hate all AI no matter what. I have even had colleagues argue with me that certain types of AI "aren't actually AI" because they have too much pride and ego to admit they use the very thing they claim to hate.
Sorry but if you can't make LLMs useful there is something wrong with how you are using them. The problwm with this whole system is that it's very helpful and valuable to the average person, as well as experts.
I'm sorry I wish you were correct but this narrative that LLMs don't work is false and bourne out of lack of experience or just flat out lies.
Lies will not defeat our very powerful opponents here, nor their massive and expensive weapons.
Pretending the enemy is useless when the opposite is true is just leaving yourself and your allies unprepared.
Yeah screw all those disabled people who benefit from it. We'll just group them in with the other counterfactuals you have to ignore to maintain an opinion like yours.
Truth. I don't support US tech either but their LLMs are state of the art and VERY useful... No point lying about it. I'm supporting Mistral.ai instead now. Their model is good enough for my use cases, and I'm hoping their recent reported uptick in users like myself helps fund the research and hardware to make them even more competitive.
Medical help. Other than that, nothing that a local LLM couldn't do.
Eventually I believe it's going to result it worse care for the most people, but in the meantime, for some people it will mean early interventions. Maybe cure diseases... What humanity does from there I don't care to guess because it will probably be worse.
Why are you so angry? I'm trying to determine if there's actually a use for these text generators, and you first say "Sorry but if you can't make LLMs useful there is something wrong with how you are using them.", and then you refuse to actually explain what you're using it for, instead making vague hand-wavey statements.
They're extremely helpful, just not at a professional level. They can help a student proof read an essay or a content creator come up with a script, but they can't help you code an app from scratch or give you a medical diagnosis.
I don't necessarily disagree with you here, I also think that no generative LLM is worth paying for, let alone a subscription with such a ridiculous price. However, I can still at least understand the appeal for a certain niche subset of people who constantly do the few stuff that a generative LLM like chatgpt excels at.
Not to mention, LLMs are very resource intensive to run. They slurp up obscene amounts of electricity and water. That's kind of a problem for a planet that's already getting boiled.
I initially subscribed to ChatGPT because I got a job as the only devops guy at an organization, when I had very limited devops experience, and ChatGPT essentially served as my mentor. I justified keeping it for a long time because it helped my productivity; bugs that I had no idea where to start with could be worked through given a few hours (or days) of back-and-forth.
As I climbed the learning curve, ChatGPT became proportionally less helpful, but I kept it because it's kind of useful for rubber ducky debugging. I did find Copilot to be pretty handy for writing docstrings (especially for keeping consistent formatting conventions), but the actual code completions were more annoying than anything.
When all was said and done, I cancelled my ChatGPT and Copilot subscriptions because I'm taking on a mortgage tomorrow and I literally just can't afford them. I have Ollama running on my homelab server, but I only have enough vRAM for a 7B-param model, and it kind of sucks ass, but whatever. At the end of the day, I like using my brain.
UPDATE (because I just thought of it after posting): I do think that "AI-as-a-mentor" is a good use-case of AI. It really helped me cut my teeth on the basics of Linux. I often find that it's easier to learn when you have a working example of code or config that you can dissect than to bash your head against the wall just trying to figure out how to get something to run at all in the first place.
For my birthday challenge this year, I'm learning how to read and write Devanagari as a surprise to my Indian grandparents. I asked my local qwen model to generate some worksheets for me to practice with, and it totally flopped. It gave away all the answers. I do think ChatGPT would have done better, but maybe I could have gotten sufficient results with a better GPU.
I know AI is an emotionally charged topic, but I think your frustration is misdirected. I find that the best way to learn tech stuff is with hand-on experience, and to that end, it works pretty well to try something, ask why it didn't work like I expected it to, and get instantaneous feedback. Or to start with a working example and pick it apart so I can learn the syntax. I'm not saying it's a replacement for reading official documentation or figuring things out for yourself, but it makes it a lot easier to get started.
Fundamentally, I'm a humanist. I believe that we should use technology in a way that augments our brain instead of circumventing it. I don't let AI write code for me, but I don't really see the harm in having it present information in a digestible format.
I've always been bored by lectures and tutorials because they're not good at meeting me at my level of experience. I don't think anyone would argue that having a tutor/mentor who gives you individual attention and meets you where you at will help you climb the learning curve way faster. And when you're in a situation where you don't have a human mentor, AI can be pretty useful.
I worked at an organization where there were no senior software people and my supervisor told me you "hey, you created this dashboard -- now deploy it". My only relevant experience was having hosted a Minecraft server on Windows 10. After a few months of iterating with ChatGPT, I knew the basics of how to use containerization and deploy an app on a RHEL server. 3 years later, I'm doing it at a tech consulting firm, and I'm the guy everyone goes to for help writing containerfiles and compose files. They promoted me from data scientist (I have an MS in data science) to solutions architect, all because I used AI to learn the basics of Linux devops, and then got a shit ton of practice by self-hosting.
I do know a few people who do. They get access to the better model, and the history is retained so it let's them build on previous stuff. Still stupid to me, because it still give you wrong info.
I "used it" for a few afternoons, to marvel what its capable of. Writing coherent short stories and shitty Weird Al songs that work surprisingly well, but the novelty soon wore off and I have no real use for it. Its like the coolest toy ever, but it can only keep me entertained for so long.
I "used it" about 6 times since then, and every time has just been to demonstrate it for old people who dont know the internet.
Boycotting ChatGPT. Have to send a message on my iPhone to my friend about it. And then I need to check my emails on my Windows computer. But I am glad I am boycotting ChatGPT.
The point is we can't compete with this saas ai products with our local models that runs on our poor homeservers. Affording a subscription is way more cheaper in shorter term.
What about using it without a subscription though ? I'm unsure whether this is good or bad for them, it loses them money but it also makes their user numbers look good so idk
At least disable "improve model for everyone" and only use temporary chats. We can't trust they'll follow it though.
Duck AI is good to anonymize your gpt session but very bad at math formatting
if you really want to hurt them use a free account and keep on asking to make you innane pictures and stuff. I mean it will waste energy though but it will cost them money.
143 Comments
LibertyLizard@slrpnk.net · 138 pts · 184d
All these boycotts I can't join since I never paid for them in the first place 😢
truthfultemporarily@feddit.org · 73 pts · 184d
You were just boycotting before it was cool.
muimota@lemmy.ml · 120 pts · 184d
Check site's social media icons, purely AI slop.
r4venw@sh.itjust.works · 58 pts · 184d
This is absurd. They couldn't spare the 3 seconds it takes to go to fontawesome to find these icons??
NikkiDimes@lemmy.world · 27 pts · 184d
The whole website was probably built by agentic AI.
criss_cross@lemmy.world · 20 pts · 184d
I’d wager the entire page was vibe coded.
angstylittlecatboy@reddthat.com · 3 pts · 183d
You can vibe code assets?
NikkiDimes@lemmy.world · 2 pts · 180d
Yup
TheGreenWizard@lemmy.zip · 55 pts · 184d
Jesus man I'm so fucking tired
XLE@piefed.social · 46 pts · 184d
Seems the site is made up of AI enthusiasts who just don't like Trump. At best, it's misguided. At worst, it's plain insincere.
They encourage you to use AI from Google and Anthropic, two companies which are evil like OpenAI.
criss_cross@lemmy.world · 17 pts · 184d
For those that don’t know Anthropic’s biggest investor is Amazon. They’re not some unknown tech company.
imetators@lemmy.dbzer0.com · 29 pts · 184d
By people my ass...
yyprum@lemmy.dbzer0.com · 5 pts · 184d
That looks like someone messed up the vector graphic icons. It doesn't look like an AI problem.
asudox@lemmy.asudox.dev · 3 pts · 184d
Oh the irony
artyom@piefed.social · 3 pts · 184d
Jesus Christ, how stupid can you be?
ToTheGraveMyLove@sh.itjust.works · 2 pts · 184d
Good god, I don't get this. That doesn't even look good.
unspeakablehorror@thelemmy.club · 54 pts · 184d
Off with their heads! GO self-hosted, go local... toss the rest in the trash can before this crap gets a foothold and fully enshitifies
mushroommunk@lemmy.today · 20 pts · 184d
LLMs are already shit. Going local is still burning the world just to run a glorified text production machine
angstylittlecatboy@reddthat.com · 2 pts · 183d
Do local LLMs really consume much more than a task like, playing a video game?
suspicious_hyperlink@lemmy.today · -15 pts · 184d
Having just finished getting an entire front end for my website, I disagree. A few years ago I would offshore this job to some third-world country devs. Now, AI can do the same thing, for cents, without having to wait for a few days for the initial results and another day or two for each revision needed
mushroommunk@lemmy.today · 33 pts · 184d
The fact you see nothing wrong with anything you said really speaks volumes to the inhumanity inherent with using "AI".
suspicious_hyperlink@lemmy.today · -14 pts · 184d
Please enlighten me. I am working on systems solving real-world issues, and now I can ship my solutions faster, with lower costs. Sounds like a win-win for everyone involved except for the offshore employees that have to look for new gigs now
Edit: I would actually rather read a reply than just see you downvoting. The point is, what you call a “glorified text generating machine”, has actual use cases
kutt@lemmy.world · 18 pts · 184d
I don’t know if it’s your fault honestly. It’s the system that makes you want to offshore your work to developing countries and not hire local employees. I get it. It’s cheaper. But when even independent developers start doing this we have reached post-late stage capitalism at this point
wholookshere@piefed.blahaj.zone · 15 pts · 184d
you missed the plot a little.
it's not that it doesn't have use cases. It burning the world down with it.
how much more water went down the drain cooling the requests you made? how about the electricity not going to local consumers but AI data centers.
all the computer components shortages...
that's still before the fact you admitted you would have hired a human, and given them food on the table instead of a corporate giant to buy another mega yacht.
suspicious_hyperlink@lemmy.today · -4 pts · 184d
Regarding the electricity not going to local customers: it’s not my fault that your country does not have appropriate regulations. None of the data centers are located where I live anyways.
Regarding hiring a human: I 100% would hire a human if there was no AI, you’re right, I’m not trying to hide it. Sucks for them I guess but I don’t see a reason why should I keep using their services if I can get a cheaper and arguably better alternative now - I’m trying to make money, not run a charity supporting development of 3rd world countries with authoritarian regimes.
However, I am pretty sure that the companies which I’m paying are currently operating on a loss with my $3/month and free coding plans. They want to grow the customer base and I’ll just switch once they will start wanting to make a profit, like the Chinese ZAI just did this week, hiking up their prices by over 3x. They are operating on a loss with my <3$/month contribution and not buying another mega yacht yet getting further away from one.
wholookshere@piefed.blahaj.zone · 3 pts · 184d
"well it doesn't effect me directly other than my bottom line, so sucks for everyone else"
if all you have to say is that, have the day you deserve.
mushroommunk@lemmy.today · 5 pts · 184d
Looks like others have come along and made my point for me while I slept. Except for calling out the dehumanizing language against the developers. They missed that one.
suspicious_hyperlink@lemmy.today · -1 pts · 184d
They weren’t my full time employees but one time gigs. I’ve worked with tons of freelancers over the years, I don’t have any special relationship with them. It’s was just a matter of whoever would offer the lowest price for the gig each time
wholookshere@piefed.blahaj.zone · 1 pts · 184d
so.... that made it okay?
Lodespawn@aussie.zone · 5 pts · 184d
link to the website please, lets see this glorious beast
VieuxQueb@lemmy.ca · 6 pts · 184d
Yeah and I wonder what are those real world products solving real world issues he talks about.
mushroommunk@lemmy.today · 1 pts · 184d
Yeah. It's not hard to say a sentence about the problem space. Heck, in good faith I'll say what I've been working on to start. I'm currently developing tools to help small communities teach their native language to younger generations as existing programs have stopped support for them.
suspicious_hyperlink@lemmy.today · 1 pts · 184d
It’s behind a login page so I’m afraid you wouldn’t see much. Also, it was never supposed to be glorious (it was not before the LLMs neither), it’s a matter of just having some form of UI as a necessity. I would be hiring actual designers if it was supposed to be a landing page or sth where the looks matter, not stick with the AI
TherapyGary@lemmy.dbzer0.com · -1 pts · 184d
Edit: I'm mocking Lodespawn VieuxQueb tbc
ch00f@lemmy.world · 5 pts · 184d
So yours and another comment I saw today got me to dust off an old docker container I was playing with a few months ago to run deepseek-r1:8b on my server's Intel A750 GPU with 8gb of VRAM. Not exactly top-of-the-line, but not bad.
I knew it would be slow and not as good as ChatGPT or whatever which I guess I can live with. I did ask it to write some example Rust code today which I hadn't even thought to try and it worked.
But I also asked it to describe the characters in a popular TV show, and it got a ton of details wrong.
8b is the highest number of parameters I can run on my card. How do you propose someone in my situation run an LLM locally? Can you suggest some better models?
lexiw@lemmy.world · 9 pts · 184d
You are playing with ancient stuff that wasn’t even good at release. Try these:
A 4b model performing like a 30b model: https://huggingface.co/Nanbeige/Nanbeige4.1-3B
Google open source version of Gemini: https://huggingface.co/google/gemma-3-4b-it
ch00f@lemmy.world · 2 pts · 184d
Any suggestions on how to get these to gguf format? I found a GitHub project that claims to convert, but wondering if there’s a more direct way.
lexiw@lemmy.world · 2 pts · 184d
ch00f@lemmy.world · 1 pts · 184d
Well, not off to a great start.
To be clear, I think getting an LLM to run locally at all is super cool, but saying "go self hosted" sort of gloms over the fact that getting a local LLM to do anything close to what ChatGPT can do is a very expensive hobby.
lexiw@lemmy.world · 1 pts · 184d
I agree, it is a very expensive hobby, and it gets decent in the range 30-80b. However, the model you are using should not perform that bad, it seems that you might be hitting a config issue. Would you mind sharing the cli command you use to run it?
ch00f@lemmy.world · 1 pts · 183d
Thanks for taking the time.
So I'm not using a CLI. I've got the intelanalytics/ipex-llm-inference-cpp-xpu image running and hosting LLMs to be used by a separate open-webui container. I originally set it up with Deepseek-R1:latest per the tutorial to get the results above. This was straight out of the box with no tweaks.
The interface offers some controls settings (below screenshot). Is that what you're talking about?
lexiw@lemmy.world · 1 pts · 183d
Those values are most of what I was looking for. An LLM is just predicting the next token (for simplicity, a word). It does so by generating every possible word with a probability associated with it, and then picking a random word from this list, influenced by its probability. So for the sentence “a cat sat” it might generate “on: 0.6”, “down: 0.2”, and so on. 0.6 just means 60%, and all the values add up to 1 (100%). Now, the number of tokens generated can be as big as the context, so you might want to pick randomly from the top 10, you control this with the parameter top_k, or you might want to discard all the words below 20%, you control this with min_p. And finally, in cases where you have a token with a big probability followed by tokens with very low probability, you might want to squash these probabilities to be closer together, by decreasing the higher tokens and increasing the lower tokens. You control this with the temperature parameter where 0.1 is very little squashing, and 1 a lot of it. In layman terms this is the amount of creativity of your model. 0 is none, 1 is a lot, 2 is mentally insane.
Now, without knowing your hardware or why you need docker, it is hard to suggest a tool to run LLMs. I am not familiar with what you are using, but it seems to not be maintained, and likely lacks the features needed for a modern LLM to work properly. For consumer grade hardware and personal use, the best tool these days is llamacpp, usually through a newbie friendly wrapper like LMStudio which support other backends as well and provide so much more than just a UI to download and run models. My advice is to download it and start there (it will download the right backend for you, so no need to install anything else manually).
ch00f@lemmy.world · 1 pts · 183d
I'll give that a shot.
I'm running it in docker because it's running on a headless server with a boatload of other services. Ideally whatever I use will be accessible over the network.
I think at the time I started, not everything supported Intel cards, but it looks like llama-cli has support form Intel GPUs. I'll give it a shot. Thanks!
Mika@piefed.ca · 2 pts · 184d
It goes down to number of vram / unified ram you have. There is no magic to make 8b perform like top tier subscription based LLMs (likely in 500b+ range, wouldn't be surprised if trillions).
If you can get to 32b / 80b models, that's where magic starts to happen.
SirHaxalot@nord.pub · 0 pts · 184d
Honestly you pretty much don't. Llama are insanely expensive to run as most of the model improvements will come from simply growing the model. It's not realistic to run LLMs locally and compete with the hosted ones, it pretty much requires the economics of scale. Even if you invest in a 5090 you're going to be behind the purpose made GPUs with 80GB VRAM.
Maybe it could work for some use cases but I rather just don't use AI.
CosmoNova@lemmy.world · 4 pts · 184d
Going local is taxing on your hardware that is extremely expensive to replace. Hell, it could soon become almost impossible to replace. I genuinely don‘t recommend it.
Even if you HAVE to use LLMs for some reason, there are free alternatives right now that let Silicon Valley bleed money and they‘re quickly running out of it.
Cancelling any paid subscription probably hurts them more than anything else.
Mika@piefed.ca · 5 pts · 184d
If LLM is tied to making you productive, going local is about owning and controlling the means of production.
You aren't supposed to run it on machine you work on anyway, do a server and send requests.
Hubi@feddit.org · 2 pts · 184d
It's not really taxing on your hardware unless you load and unload huge models all day or if your cooling is insufficient.
sudoer777@lemmy.ml · 2 pts · 184d
Last I checked it costed ~$6000 to run a high end LLM at terrible speeds and that was before the RAM price hike. And at the rate things are changing it might be obsolete in a few year. And I don't have that much money either.
I'm going to stick with the free OpenCode Zen models for now and maybe switch to OpenCode Black or Synthetic or whatever when they stop being available and use the open weight models there for now.
ZILtoid1991@lemmy.world · -2 pts · 184d
I would, if I found even a remotely good usecase for LLMs. Would be useful for contextual search on a bunch of API documentation and books on algorithms, but I don't want a sychophantic "copilot" or "assistant", that does job so bad I would be fired for, all while being called ableist slurs and getting blacklisted from the industry.
unspeakablehorror@thelemmy.club · 1 pts · 183d
I only have a need for RAG and it works well, wouldn't say I need w copilot either. Everyone probably has some use case for this stuff but it's gonna be different for everyone, the basic llm conversation model it's really just a stepped up interactive Google
canofcam@lemmy.world · 49 pts · 184d
not just chatgpt.
Stop using Amazon, Meta products, Netflix, Spotify, etc.
All of these massive corporations are actively making life worse for people, and it will only get worse and worse as people continue to stay subscribed.
The only option is to log off and find an alternative.
SlimePirate@lemmy.dbzer0.com · 18 pts · 184d
Google ?
msage@programming.dev · 17 pts · 184d
Microsoft, Tesla, Twitter, Nvidia, Nestle...
... in fact, forget buying altogether.
dovahking@lemmy.world · 1 pts · 183d
You will owe nothing and you'll be happy.
msage@programming.dev · 2 pts · 182d
I will own my self-hosted open-sourced software and data.
Fuck digital overlords.
canofcam@lemmy.world · 4 pts · 184d
Yep, should go without saying! I didn't list every company but there are plenty more who belong there.
XLE@piefed.social · 6 pts · 184d
Apparently, they just took somebody else's idea and made it all about ChatGPT:
Resist and Unsubscribe encourages you to get away from Google and Amazon, as well as OpenAI. QuitGPT endorses both (Google directly, Amazon indirectly).
Resist and Unsubscribe is a holistic project; QuitGPT cuts out everything except for one product.
Resist and Unsubscribe doesn't inadvertently promote a single product; QuitGPT practically functions as a sneaky advertising campaign (kind of like how Larry David said he wouldn't invest in FTX).
It seems pretty clear which project is better, even if I don't agree with everything the guy behind it said:
I don't think voting with your dollars will make a huge difference when every ChatGPT subscription costs OpenAI money, but go off I guess
UnderpantsWeevil@lemmy.world · 37 pts · 184d
Subscriptions? You idiots are paying for this shit?
bridgeenjoyer@sh.itjust.works · 13 pts · 183d
Businesses do and try to get employees to use it... Stupid
sfgifz@lemmy.world · 7 pts · 183d
Well the fucked up part is businesses also want to see how well their "investment" AI is doing, which means tracking how much employees use the AI and other cooked up metrics. You can't get away from it no matter how irritating it makes your job and life.
bridgeenjoyer@sh.itjust.works · 4 pts · 183d
Its infuriating because now every single meeting is "BLARGH HOW CAN WE MAKE DA AI DO THIS? HOW CAN WE UTILIZE THE AI???!!!"
makes me want to scream.
myfunnyaccountname@lemmy.zip · 1 pts · 183d
All the time at work. Free tier and paid. Translate shit from foreign customers. Extract text from screen shots. There’s lots of uses, just like any other tool. But it’s a tool, it’s not fucking magic or the devil. Just a tool.
FlashMobOfOne@lemmy.world · 35 pts · 184d
Unsubscribe from all the things.
PhoenixDog@lemmy.world · 16 pts · 184d
Except my VPN. You don't need to know where I live. And my IPTV.
FauxLiving@lemmy.world · 7 pts · 184d
Unless you're (not you PhoenixDog specifically, the general You referring to the reader) blocking your browser fingerprinting, using Ad blocker, DNS filtering from trusted sources, not using social media app, not installing apps that require Google Play Services and not using your credit card for digital payments then simply changing the IP address that the server sees doesn't do very much.
VPNs can be part of a solution but if you are not doing all of the other things then they do not hide who you are because of the plethora of other avenues that you can be tied to an existing advertising identity.
queueBenSis@sh.itjust.works · 7 pts · 184d
but be careful of which VPN provider you’re using. Mulvad is a good option
PhoenixDog@lemmy.world · 2 pts · 184d
Using Nord right now. Looking into that one though I may switch.
FlashMobOfOne@lemmy.world · 6 pts · 184d
Yeah, VPN's are unfortunately essential. If I hadn't made it a project to request deletion of my info off the top 30 background check sites, I'd be using something like DeleteMe too.
SaharaMaleikuhm@feddit.org · 11 pts · 184d
I just unsubscribed from my landlord
Jankatarch@lemmy.world · 34 pts · 184d
Educational institutions are buying microsoft services as a trend and we need them to boycott also.
Getting bus pass in my university takes 5 business days so they can "save cost on students who don't use it."
Meanwhile everyone's student gmail implicitly has a subscription to copilot, gemini, and openai at the same time; it doesn't matter if you are using them or not.
Paid by our tuition no doubt.
TankovayaDiviziya@lemmy.world · 30 pts · 184d
People pay for subscription to ask AI?
SaharaMaleikuhm@feddit.org · 12 pts · 184d
Or worse: have it make images that they then post online
mynona@lemmy.world · 2 pts · 184d
The piss filter burns so badly. If people post those, I appreciate them volunteering to be blocked.
humanspiral@lemmy.ca · 20 pts · 184d
Grok/Mechahitler owner likely has donated more to GOP/Trump rule. Anthropic CEO wrote an essay that outlined dangers of AI recently, but also underlined how the pure good US must develop Skynet to oppress all of the bad evil opposition to the US empire.
The entire industry is lining up to get US government military/surveillance applications to use the absurd level of datacenters/energy that are being committed to, because "Skynet now or China wins" is even more aggressive consensus than "too big to fail".
Jeremyward@lemmy.world · 10 pts · 184d
Yes agree, i have quit MSFT fully (switched to Ubuntu), Amazon prime is gone and no longer buying from them. Cancelled my Claude subscription, my gmail now auto forwards to my local server, switched my web hosting from AWS to Hetzner, Cancelled my netflix, back to library and jellyfin local server. Same with Spotify. No more Safeway or QFC - Costco is good enough and has the lowest prices locally, or occasionally Trader Joes. Its saving me a whole shit ton of money.
eletes@sh.itjust.works · 19 pts · 183d
If you thought the banks were too big to fail, get ready for this fucking bailout
Suavevillain@lemmy.world · 18 pts · 184d
I don't pay for AI. There is no real value in it at massive scale for it to be paid for me. Plus it is harmful. I can local host AI for small tasks like cleaning up some data.
Gatsby@lemmy.zip · 8 pts · 184d
Yep! I feel way, way safer when I use something running on my laptop as opposed to feeding things into the bottomless hole that is Microsoft or Anthropic.
All I want is something to double-check the verbiage of an email to make sure I'm not coming off like an asshole.
Scrollone@feddit.it · 3 pts · 183d
Also, why paying for AI? You can use it for free, and when you finish your free daily limit, you just switch to another AI: Gemini, Perplexity, Claude, Le Chat, DeepSeek... rinse and repeat.
UltraGiGaGigantic@lemmy.ml · 16 pts · 184d
People pay for this shit?
ssfckdt@lemmy.blahaj.zone · 16 pts · 183d
Honestly, people are probably going to start using it much less once they put in ads on the the low and free tiers.
They're realizing they can't turn profit over expenses when the VC money dries up without jacking price.
Same shit different fad. Offshoring was all the rage until the demand meant they started charging more, the quality went down, and the reality of cross-timezone management became obvious. Cloud was all the rage until the prices went up from demand and also kept running into outages, and now some companies are "re-inhousing" their infrastructure.
Cant' wait till they realize they also can't deliver at regular clip with only half the workers. Wont' be soon enough, alas.
Azrael@reddthat.com · 10 pts · 184d
I've been a long time user and subscriber of ChatGPT plus. It has helped me tremendously and I probably wouldn't be where I am without it.
I canceled mt subscription recently. I won't continue to give money to ClosedAI. There are alternatives.
Sam Altman can burn in hell.
innermachine@lemmy.world · 4 pts · 184d
Mind if I ask how it's helped you along? My lil bro used it to do basically all of his college work LMFAO. He also uses ai screening tools to make sure he changes shit enough that the professor doesn't catch it, well one day he screened a assignment from the professor and it was flagged as AI produced LMFAO
aceshigh@lemmy.world · 3 pts · 184d
I have memory issues (it’s a brain issue) and struggle to do some basic things, but higher order thinking is natural for me. So ChatGPT does the things I struggle with, I review it and then use that information for my bigger goals. I got a lot done last year because of chatgpt, and rn I’m on the last leg of the project. I spent a few years trying to complete the project alone and didn’t get far because of the memory issues.
How ai is used is dependent on the user itself. Some people just don’t want to put effort like your brother, some people have disabilities or other issues that prevent them from doing bigger things they’re otherwise capable of doing.
Katana314@lemmy.world · 3 pts · 184d
For honesty, recent example from me:
I bought about a dozen epub comics. They were formatted with a hardcoded 600x450 width or so, maybe expecting a particular device. Having recently worked with epubs to format my own (word) book, I knew the format, and basically wanted to use Python standard library tools to unzip them, rip out some useless sizing/styling code (from hundreds of XHTML files), and zip them back up.
I hadn't used Python professionally in a few years, so this was an annoying back and forth to work out the process and remind myself of syntax, especially considering this was something I was just doing for a few of my own books. Instead, taking every important piece of this puzzle/process I'd researched, I instead described the problem to ChatGPT, specifically pointing it to the Python standard libraries I wanted to use. It gave me a one-page program that was mostly complete and I only needed to change in a few areas.
I don't think I'd ever pay money to AIs for a variety of reasons. I take that assistance as it comes, and could live without it.
Azrael@reddthat.com · -1 pts · 184d
I personally use AI to help me with my work. I used to work in an IT repair facility where I would use an LLM to help me troubleshoot problems. For example if a computer won't turn on and the motherboard is flashing and beeping, asking an LLM is 10x faster than having to use the internet to hunt for a user manual of one specific model of computer from 40 years ago, made by a brand which no longer exists.
Now I work in Computer Games Development after getting my degree. AI is extremely useful for helping me come up with ideas, making prototypes, and even generating small amounts of code in C++ and C#, which is what I mainly use. Games Dev isn't difficult. I can do it myself. But it can be very tedious. That's why AI is so helpful.
As for what your little brother does, i've experienced that. People generating code and having no idea what it does. It might help you pass your subjects in school, but it's only gonna screw you over in the long run.
I don't mind people using AI to help them with their work as long as they don't depend on it.
ToTheGraveMyLove@sh.itjust.works · 5 pts · 184d
You're literally using AI to think for you and then pretending like you're not dependent on it...
BCsven@lemmy.ca · 3 pts · 184d
I don't understand the down votes on this comment. What you have said seems reasonable
Azrael@reddthat.com · 2 pts · 184d
Probably because it challenges the narrative of people who automatically hate all AI no matter what. I have even had colleagues argue with me that certain types of AI "aren't actually AI" because they have too much pride and ego to admit they use the very thing they claim to hate.
Dyskolos@lemmy.zip · 10 pts · 184d
stressballs@lemmy.zip · 10 pts · 184d
Sorry but if you can't make LLMs useful there is something wrong with how you are using them. The problwm with this whole system is that it's very helpful and valuable to the average person, as well as experts.
I'm sorry I wish you were correct but this narrative that LLMs don't work is false and bourne out of lack of experience or just flat out lies.
Lies will not defeat our very powerful opponents here, nor their massive and expensive weapons.
Pretending the enemy is useless when the opposite is true is just leaving yourself and your allies unprepared.
Dyskolos@lemmy.zip · 5 pts · 184d
stressballs@lemmy.zip · -1 pts · 184d
Yeah screw all those disabled people who benefit from it. We'll just group them in with the other counterfactuals you have to ignore to maintain an opinion like yours.
Dyskolos@lemmy.zip · 0 pts · 184d
BCsven@lemmy.ca · 3 pts · 184d
They are a tool, but some people use it as a replacement for actual knowledge, and take it as gospel. Its going to create a generation of idiots
Darkmoon_AU@lemmy.zip · 2 pts · 184d
Truth. I don't support US tech either but their LLMs are state of the art and VERY useful... No point lying about it. I'm supporting Mistral.ai instead now. Their model is good enough for my use cases, and I'm hoping their recent reported uptick in users like myself helps fund the research and hardware to make them even more competitive.
merc@sh.itjust.works · 1 pts · 183d
What is it that you find useful?
stressballs@lemmy.zip · -1 pts · 183d
Medical help. Other than that, nothing that a local LLM couldn't do.
Eventually I believe it's going to result it worse care for the most people, but in the meantime, for some people it will mean early interventions. Maybe cure diseases... What humanity does from there I don't care to guess because it will probably be worse.
merc@sh.itjust.works · 1 pts · 183d
What do you mean by "medical help"?
stressballs@lemmy.zip · -2 pts · 183d
Go that way and keep going until you see the sign marked none of your fucking business.
merc@sh.itjust.works · 2 pts · 183d
Why are you so angry? I'm trying to determine if there's actually a use for these text generators, and you first say "Sorry but if you can't make LLMs useful there is something wrong with how you are using them.", and then you refuse to actually explain what you're using it for, instead making vague hand-wavey statements.
Gorilladrums@lemmy.world · 1 pts · 184d
They're extremely helpful, just not at a professional level. They can help a student proof read an essay or a content creator come up with a script, but they can't help you code an app from scratch or give you a medical diagnosis.
Dyskolos@lemmy.zip · 1 pts · 184d
Gorilladrums@lemmy.world · 1 pts · 184d
I don't necessarily disagree with you here, I also think that no generative LLM is worth paying for, let alone a subscription with such a ridiculous price. However, I can still at least understand the appeal for a certain niche subset of people who constantly do the few stuff that a generative LLM like chatgpt excels at.
Dyskolos@lemmy.zip · 1 pts · 184d
gravitas@pie.gravitywell.xyz · 9 pts · 184d
Or better yet Unsubscribe but dont stop using ths service, if enough go from paying customer to wasteful freeloader they will lose even more.
sefra1@lemmy.zip · 12 pts · 184d
Not sure if it works like that.
Just from using their platform you're giving them value, at least in the form of data.
OneWomanCreamTeam@sh.itjust.works · 14 pts · 184d
Not to mention, LLMs are very resource intensive to run. They slurp up obscene amounts of electricity and water. That's kind of a problem for a planet that's already getting boiled.
chiliedogg@lemmy.world · 1 pts · 184d
What we need to do then is poison the well and provide bad training data.
biofaust@lemmy.world · 2 pts · 184d
One seahorse emoji a day.
Cruxifux@feddit.nl · 9 pts · 184d
You can subscribe to chatGPT?
Dojan@pawb.social · 22 pts · 184d
Yes. I think it’s like $20 a month.
--
Edit: LMAO so I was fuck-off wrong. It's $10, $30, and $280 per month. At least in my currency (Swedish Crowns).
Don't use the stochastic parrot, and definitely don't fucking shell out 280 a month for it. Holy fuck.
capuccino@lemmy.world · 2 pts · 184d
Mexican pesos:
Edit: I was reading through their plans and I didn't knew that Sora was a pay service. Now it's easily to know who paid.
sudoer777@lemmy.ml · 1 pts · 184d
For coding $280/mo is peanuts compared to how much the Claude API costs
Dojan@pawb.social · 2 pts · 184d
Right, I also like paying to not know what the heck is going on in the code I maintain.
Jakeroxs@sh.itjust.works · 1 pts · 184d
It is 20 in the US
Jaysyn@lemmy.world · 9 pts · 184d
It blows my mind people actually subscribe to mostly wrong answers.
Bwaz@lemmy.world · 9 pts · 184d
I kinda wish I'd been using it so I could quit!
sircac@lemmy.world · 7 pts · 184d
Any reference to Trump's donors to back that Gepeto is the biggest one? I would like to see the top 10 or 100 list...
epicshepich@programming.dev · 4 pts · 184d
I initially subscribed to ChatGPT because I got a job as the only devops guy at an organization, when I had very limited devops experience, and ChatGPT essentially served as my mentor. I justified keeping it for a long time because it helped my productivity; bugs that I had no idea where to start with could be worked through given a few hours (or days) of back-and-forth.
As I climbed the learning curve, ChatGPT became proportionally less helpful, but I kept it because it's kind of useful for rubber ducky debugging. I did find Copilot to be pretty handy for writing docstrings (especially for keeping consistent formatting conventions), but the actual code completions were more annoying than anything.
When all was said and done, I cancelled my ChatGPT and Copilot subscriptions because I'm taking on a mortgage tomorrow and I literally just can't afford them. I have Ollama running on my homelab server, but I only have enough vRAM for a 7B-param model, and it kind of sucks ass, but whatever. At the end of the day, I like using my brain.
UPDATE (because I just thought of it after posting): I do think that "AI-as-a-mentor" is a good use-case of AI. It really helped me cut my teeth on the basics of Linux. I often find that it's easier to learn when you have a working example of code or config that you can dissect than to bash your head against the wall just trying to figure out how to get something to run at all in the first place.
For my birthday challenge this year, I'm learning how to read and write Devanagari as a surprise to my Indian grandparents. I asked my local qwen model to generate some worksheets for me to practice with, and it totally flopped. It gave away all the answers. I do think ChatGPT would have done better, but maybe I could have gotten sufficient results with a better GPU.
bold_atlas@lemmy.world · 0 pts · 183d
Sorry but thinking of a shitty chatbot as your "mentor" is absolute brainrot.
epicshepich@programming.dev · 2 pts · 183d
I know AI is an emotionally charged topic, but I think your frustration is misdirected. I find that the best way to learn tech stuff is with hand-on experience, and to that end, it works pretty well to try something, ask why it didn't work like I expected it to, and get instantaneous feedback. Or to start with a working example and pick it apart so I can learn the syntax. I'm not saying it's a replacement for reading official documentation or figuring things out for yourself, but it makes it a lot easier to get started.
Fundamentally, I'm a humanist. I believe that we should use technology in a way that augments our brain instead of circumventing it. I don't let AI write code for me, but I don't really see the harm in having it present information in a digestible format.
I've always been bored by lectures and tutorials because they're not good at meeting me at my level of experience. I don't think anyone would argue that having a tutor/mentor who gives you individual attention and meets you where you at will help you climb the learning curve way faster. And when you're in a situation where you don't have a human mentor, AI can be pretty useful.
I worked at an organization where there were no senior software people and my supervisor told me you "hey, you created this dashboard -- now deploy it". My only relevant experience was having hosted a Minecraft server on Windows 10. After a few months of iterating with ChatGPT, I knew the basics of how to use containerization and deploy an app on a RHEL server. 3 years later, I'm doing it at a tech consulting firm, and I'm the guy everyone goes to for help writing containerfiles and compose files. They promoted me from data scientist (I have an MS in data science) to solutions architect, all because I used AI to learn the basics of Linux devops, and then got a shit ton of practice by self-hosting.
imetators@lemmy.dbzer0.com · 4 pts · 184d
Wait. Normal people who use chatgpt just for basic questions are paying for it?
BCsven@lemmy.ca · 3 pts · 184d
I do know a few people who do. They get access to the better model, and the history is retained so it let's them build on previous stuff. Still stupid to me, because it still give you wrong info.
DarrinBrunner@lemmy.world · 4 pts · 184d
I've never used it. I play around with a local installation sometimes, that's the extent of my experience with AI.
emmy67@lemmy.world · 4 pts · 184d
Quit? Only a fool would waste their time on it.
atropa@piefed.social · 4 pts · 184d
Mika@piefed.ca · 3 pts · 184d
Corporate would still use it 😒
blorgnis@lemmy.world · 3 pts · 184d
Never paid for it, but I would use it here and there. Never again, account deleted.
Xylian@lemmy.world · 4 pts · 184d
You can continue using it for free, it makes cost for them ;)
blorgnis@lemmy.world · 5 pts · 184d
Well, I like that, but I also like the idea of not helping them train their model
yermaw@sh.itjust.works · 2 pts · 184d
I "used it" for a few afternoons, to marvel what its capable of. Writing coherent short stories and shitty Weird Al songs that work surprisingly well, but the novelty soon wore off and I have no real use for it. Its like the coolest toy ever, but it can only keep me entertained for so long.
I "used it" about 6 times since then, and every time has just been to demonstrate it for old people who dont know the internet.
stressballs@lemmy.zip · 3 pts · 184d
It might be helpful to the movement if memes kike this weren't blatantly untrue.
zr0@lemmy.dbzer0.com · 2 pts · 184d
Boycotting ChatGPT. Have to send a message on my iPhone to my friend about it. And then I need to check my emails on my Windows computer. But I am glad I am boycotting ChatGPT.
ruvanoit@lemmy.world · 2 pts · 183d
The point is we can't compete with this saas ai products with our local models that runs on our poor homeservers. Affording a subscription is way more cheaper in shorter term.
demonsword@lemmy.world · 2 pts · 183d
this won't be true anymore when the AI bubble bursts
ruvanoit@lemmy.world · 1 pts · 181d
In my perspective as a cs graduate, it'll only have a decline instead of a pop. We've already exceeded the point of no return.
termaxima@slrpnk.net · 1 pts · 184d
What about using it without a subscription though ? I'm unsure whether this is good or bad for them, it loses them money but it also makes their user numbers look good so idk
SlimePirate@lemmy.dbzer0.com · 2 pts · 184d
At least disable "improve model for everyone" and only use temporary chats. We can't trust they'll follow it though. Duck AI is good to anonymize your gpt session but very bad at math formatting
Paddy66@lemmy.ml · 1 pts · 166d
While people are at it they could quit ALL big tech. Here's our site (that was not AI generated!):
https://www.rebeltechalliance.org/stopusingbigtech.html
(Please share with your friends and family - we have zero marketing budget - thank you!)
super_user_do@feddit.it · 1 pts · 184d
I dont give them any valuable information about me and I never pay. I also only use it in the browser and not even that often
HubertManne@piefed.social · 0 pts · 184d
if you really want to hurt them use a free account and keep on asking to make you innane pictures and stuff. I mean it will waste energy though but it will cost them money.
Octagon9561@lemmy.ml · -1 pts · 183d
I just use Chinese AI atp. Deepseek and Kimi.
Beep@lemmus.org · 0 pts · 183d
Glm also?
MasterNerd@lemmy.zip · -3 pts · 184d
Nah I'm gonna use my free account to prompt a bunch if inane shit to drive up operating costs while poisoning their training data