Hello sailors,
I have been job hunting for a while and I have felt a great disadvantage in my job search due to my lack of access to high-quality LLMs. Writing cover letters is honestly so bullshit. GPT-3 is honestly quite bad nowadays, but as a true pirate at heart I couldn’t quite get myself to cough up the coin for OpenAI’s GPT4 out of principle. I hate them for putting their cutting edge technology behind a paywall, making it inaccessible for their own gain. I feel like this is not what the internet was supposed to be. So today, call me the great emancipator cuz i’m teaching u how to get that shi for free baby
Requirements: Docker
It’s all gonna be based off of this github repo: gpt4free
Installing through docker (there’s also a way to install with Python PIP if that’s more convenient for you. The docker worked for me though)
-
docker pull hlohaus789/g4f -
docker run -p 8080:8080 -p 1337:1337 -p 7900:7900 --shm-size="2g" -v ${PWD}/hardir:/app/hardir hlohaus789/g4f:latest -
Open up the webui in your browser at localhost:8080
-
In the “Provider” dropdown in the bottom look for “Liaobots”
-
Choose “gpt-4-plus” under the “Models” dropdown
??? Profit
The cool thing about gpt4free is that there’s a lot of providers and a lot of models to choose from! So if gpt-4-plus from Liaobot doesn’t work for you you can switch to something else easily. Do note that some models require you to provide an authentication token or be logged in. Most of them work right out of the box tho.
*this post was not made with any use of an llm I promise ;)
^^list of gpt4 providers
You must log in or # to comment.
This feels… Scammy? Not to be accusatory, but gpt 4 is expensive to run. It is impossible for people to use it for free.
What llm is actually providing the response here? Either someone is footing the bill for an API and acting as a proxy, a situation which raises many red flags, or the model you’re talking to is something far cheaper to run, like a mistral model.
Even the second case is sketchy. 😅
Well these guys are probably hoovering up all your data so maybe it’s viable if people are sending them saucy information
Figuring this one out myself too. Liaobots is a chinese site and when asked about the
gpt-4-plusmodel it mentions that it’s based on the gpt-4 architecture but i’m not too sure as well. My guess is that the chinese reverse engineered gpt-4 and put it out there, like a bootleg gpt-4.Bing tends to get heavily ratelimited.
So maybe, yes you’re there’s a sketchiness and too-good-to-be-true factor to this so maybe calling it “free gpt-4” is a bit misleading. But it’s still super cool to have an all-in-one interface for all your llm providers and models :)
There is almost no chance that it is truthfully based on gpt 4. If you want a free, open source llm with 32k context and generous limits, I recommend using huggingface.co/chat/
The nous-hermes model (you can select different models) is uncensored, and performs really well for a open source model. Plus, they have data controls so you can turn off data gathering per model. Huggingface is a reputable vendor, and doesn’t claim to be something it isn’t.
Huggingface is included in the providers listed here actually :) and i agree. I can’t read chinese so i don’t know how legit or sketch liaobots.com is
I mean, be careful. These llms can be honeypots for data. Like, if you’re using it for cover letters, or work, you’re sending tons of personal info to random websites.
I would recommend sticking to actual, reputable vendors for llms, or running your own. I have a GTX 1070 and can run some pretty decent models these days locally using koboldai.
Bing is probably the only way to use gpt 4 without paying for it, and Microsoft probably won’t steal your bank account info.
Good advice 👍
AI Horde provides LLMs for free. They’re not GPT4, but they’re uncensored and crowdsourced by the community. There’s also not tracking or ads. Here’s a frontend for LLMs. No need to install anything either. If you have a decent GPU, you can also assist others with generations
Thanks for this 🙌 this is also pretty good
Oh, man. I hope you won’t take this poorly, but I can promise you that setting up some kind of GPT4-based LLM to wrote cover letters won’t help you find another job. Not ever.
I’m not sure if you’re trolling (if you are, LOL), but this is not the tree to bark up. Good luck though. The job market sucks at the moment.
Cover letters are about showing you are willing to put the effort in to writing a cover letter, nobody actually reads cover letters. LLMs are great for them until companies realize everyone is just using LLMs.
They already know that and have since people started using LLMs.
Also, if you write a cover letter with an LLM and don’t tweak it, it’s obvious to anyone who knows what to look for.
My sister got a job with a primarily LLM-written cover letter. When trying to sway someone to your side, how you say something can matter more than what you say.
For people who aren’t good at articulating themselves, noting down key dot points about their skills and job history alongside a job description, then asking a LLM to write out a cover letter can be very helpful, even if only to get a rough draft.
As an aside, out of curiosity, I asked ChatGPT 3.5 to rewrite my comment above and got the following, not sure which is better TBH.
My sister landed a job using a cover letter mostly written by a language model. When persuading someone, the tone and style of your message can outweigh its content.
For individuals who find it challenging to articulate themselves, outlining key skills and job experiences alongside a job description, and then seeking assistance from a language model to compose a cover letter, can prove highly beneficial, even if only as a starting point.
It would sound snarky to say
correlation is not causationso let me say instead: just because your sister got a job with that cover letter doesn’t mean it’s the reason.Hiring managers and HR don’t always read cover letters. Most never do. All your applications go into a applicant tracking system (ATS) anyway and your info gets extracted. Lately, folks uses LLMs to query that stuff: “find me all the candidates who went to top colleges,” or “which of these candidates can bend a steel beam with their bare hands,” etc.
It’s not the right sub for this, but spending a bunch of energy on cover letters won’t do much. Better cover letters won’t hurt and BAD cover letters might hurt, but there are many other reasons someone doesn’t get call-backs.
I have zero clue what field OP works in, but they’re actually free to DM me for advice if they like.
I have been job hunting for a while and I have felt a great disadvantage in my job search due to my lack of access to high-quality LLMs. Writing cover letters is honestly so bullshit.
Has job hunting gotten so bad that it’s just LLMs reading the output of other LLMs?
It’s pretty bad out there
Yeah basically, people are using AI to write applications and cover letters, and recruiters are using AI to read and filter them, so it’s just robots talking to each other.
I’m heard some other horror stories too, like companies requesting a “one way remote interview” which basically means they send you a list of questions, and you’re supposed to record a video of yourself answering them as if you were in a proper interview and then send it to them.
At which point I’d rather be homeless personally, but that’s maybe just me lol
I would be careful with some of these providers depending on your usage.
You are potentially sending a ton of info to them…I have access to Bing Chat Enterprise through my company, and only because its the Enterprise version i am half confident in using it with more restrictive data.
Though the frontend of copilot is so heavy and sucks, so i have a proxy for GPT-API to Bing Chat.
Had hoped GPT4All Bing provider would support login, but sadly not, so essentially had to reimplement it all myself.Seems scammy (not you, but the “free” gpt4 providers)
Also check out https://gpt4all.io/ if you don’t want to use Docker (for some reason).
I see that it uses a local model. How good is it really? How fast are the responses in consumer grade gpus?
It let’s you pick from a whole bunch of local models to download, some trained by Microsoft and the like. In my experience it’s pretty responsive on a 2070s ive had for years now, but the responses aren’t as good as something like gpt4. Probably about on par with gpt3 in most cases if you choose a larger model.
LMStudio is excellent for downloading and running local models. Performance and quality are vastly improved since the first versions of LLMs came out. I get near instant responses on my laptop with only a GTX1650 in it.
With that said, they still fail to produce much of value. YMMV
This is just free local models right? Not models from OpenAI.
No these are not local models. It’s an all-in-one front end for different LLM providers and models. Although in the documentation you can specify your own local model
Yeah, so just free models available, unless you already bought the paid ones. I fail to see the piracy part, or the “free chatgpt”. You’re just advertising that there are free alternatives, which I don’t mind particularly, but it just came of as something else.
deleted by creator
I might self-host it and provide the mirror for friends and stuff. I’ll let you know when the motivation strikes me and i’ll send u a link :-)
You could use termux and run it on an android phone.
deleted by creator
OP, is this online or will I need a big CPU with lots of storage to run this?
online :)
I’m going to be on the job hunt soon, this will be helpful.
If you really want to use gpt-4 for a cover letter (don’t think it’ll help that much, but I won’t judge), just get yourself an API Key and use the API with a local frontend. The requests you need for a cover letter will likely amount to a few cents.
I get the whole pirate spirit thing, but this all seems wonky at best. If you want fully free gpt-4, there’s always Bing Chat.
I use the android app and it has a GPT-4 option… Am I missing something? Isn’t it already free?
Tried out GPT4 when it first dropped. Glorified chatbot that has zero agency. I struggle to find an application for it other than summarizing blocks of text.
