How to use Llama 3(70B) API for FREE (beats GPT4 for business!)

Поделиться
HTML-код
  • Опубликовано: 2 янв 2025

Комментарии • 74

  • @Dream22121
    @Dream22121 Месяц назад +1

    Short,Conscise and perfect. This is one of the best videos on the topic!

  • @DAvgGamer98
    @DAvgGamer98 7 месяцев назад +8

    Excellent video. Suggesting to optimize the video more in terms of description and keywords so that you end up getting more search traffic. Please keep up the good work.

    • @aiunleashed509
      @aiunleashed509  7 месяцев назад +1

      Thank you, and thanks for the great suggestion. You are right I will add more to the description

  • @francescoavallone-xg7vt
    @francescoavallone-xg7vt 32 минуты назад

    Thank you very much, I saved a lot of my time with this video. Very very good!

  • @GurbyTheGreat
    @GurbyTheGreat 8 месяцев назад +10

    If it's free your prompt content is being saved. 70b costs lots of money to run on a server like this, it's not free!

    • @aiunleashed509
      @aiunleashed509  8 месяцев назад +14

      You make a good point, nothing is truly free. However, I still think what meta is doing by open sourcing a top tier model is a big deal. It allows the entire universe of cloud providers to host llama, not just being stuck with a few big tech companies gatekeeping. Also puts downward pressure on AI API pricing in general.

    • @jonahsatheesh5346
      @jonahsatheesh5346 6 месяцев назад +6

      So what if the prompts are saved, it just makes the model better

    • @jurassicthunder
      @jurassicthunder 5 месяцев назад

      ​@@jonahsatheesh5346data is the new oil. your personal data is actually worth money. you are trading it to use the product.

    • @thecompanioncube4211
      @thecompanioncube4211 2 месяца назад

      Agreed. But for learning purpose it's good. For actual enterprise application your company will bear cost of hosting of Api

  • @LucaTV296
    @LucaTV296 3 месяца назад +1

    it says the term export is not recognized can some one help me? untile 1:42 everything was working

    • @aiunleashed509
      @aiunleashed509  2 месяца назад

      Hey sorry for the late reply. My best advice is use cursor AI code editor and put the error you get in the chat. It will likely fix it right away

    • @LucaTV296
      @LucaTV296 2 месяца назад

      @@aiunleashed509 ok im currently not home but when i will get home i will write it

    • @VishwanthD
      @VishwanthD Месяц назад

      export is used for bash and zsh shells which are in mac. You might probably be in windows powershell so use the command $env:GROQ_API_KEY="YOUR_API_KEY"

  • @mohakgupta3749
    @mohakgupta3749 Месяц назад +1

    Can I use it for langchain developing RAGs with chat history, not providing sensitive information?

    • @aiunleashed509
      @aiunleashed509  Месяц назад

      Yes you can use it in Langchain. Not sure exactly what your doing, but think of if as a free alternative to using OpenAI GPT4o paid credits

  • @Ajaysharma-ox9rr
    @Ajaysharma-ox9rr Месяц назад +1

    when i am executing it as simple function it works
    but creating an API via express its giving 401 invalid token error ?

    • @aiunleashed509
      @aiunleashed509  Месяц назад

      Hey, I would try loading the relevant files into Cursor composer, and pasting in the error.

  • @paras6061
    @paras6061 8 дней назад

    Finally i found quality video to find free api key of llama ,that's great & thank you sir

  • @mohamedalichakroun6967
    @mohamedalichakroun6967 4 месяца назад +1

    Thanks , but It can extract data from images ?
    I tried llava but it gave me a bad result
    Always say that the image provided doses not have good resolution ?
    Can you please make a video about that like you did with gpt4 for conversion img to json

    • @aiunleashed509
      @aiunleashed509  4 месяца назад

      I don't think you can use this method for AI vision. Working on a video about AI vision with Claude

  • @play-good
    @play-good 20 дней назад +3

    what is the limit??

    • @aiunleashed509
      @aiunleashed509  8 дней назад +1

      console.groq.com/settings/limits
      for most people with free account llama 3.3 will be 30 requests per minute, 1000 per day

    • @play-good
      @play-good 8 дней назад

      @@aiunleashed509 thx u

  • @emilsmezinskis7905
    @emilsmezinskis7905 5 месяцев назад +2

    why I have this error? export GROQ_API_KEY="my_api_key"
    export: The term 'export' is not recognized as a name of a cmdlet, function, script file, or executable program.
    Check the spelling of the name, or if a path was included, verify that the path is correct and try again.

    • @aiunleashed509
      @aiunleashed509  5 месяцев назад +3

      its likely you are trying to run export in a windows environment. Try set GROQ_API_KEY=my_api_key...google setting windows environment variables if that doesn't work

    • @thecompanioncube4211
      @thecompanioncube4211 2 месяца назад +1

      You're in windows. Export command is for linux/macos. Google the ways

  • @knight5970
    @knight5970 8 месяцев назад +4

    That's great. Thanks for the video, sir.

  • @Shiffo
    @Shiffo 3 месяца назад +1

    Is your data secure if you run it through Groq? i want to use llama on excel data but don’t want to leak information.

    • @aiunleashed509
      @aiunleashed509  3 месяца назад

      personally, I use Groq for more proof of concept type projects, and do not send sensitive data. A key note from Groq:
      Notwithstanding the restriction on licensing set forth in section 4, Content Available through the Services, User Data shall be owned by you and is not retained or used by Groq other than to perform the Generative AI Services hereunder.
      Also, groq.com/security/

    • @leonardopo6202
      @leonardopo6202 Месяц назад

      did you find a solution for your issue? I am also looking to find a way to use llama models on some google sheets.

    • @Shiffo
      @Shiffo Месяц назад

      @ i did not yet, it seems that llama works well with csv files but excel is a bit of drama. You could look into gemini since that is integrated with google sheets, but might be less secure for your data

  • @definitelynorandomvideos24
    @definitelynorandomvideos24 8 месяцев назад +1

    Any idea what the rate limits are? Couldn't find that information on google :/ I only saw the max tokens per request.

    • @aiunleashed509
      @aiunleashed509  8 месяцев назад +3

      Yes the limits for llama3-70b-8192 are 30 requests per minute & 6,000 tokens per minute. Also max 14,400 requests per day.

    • @ianbraganza3065
      @ianbraganza3065 7 месяцев назад

      How much data can be transferred per request? I'm a newbie

    • @aiunleashed509
      @aiunleashed509  7 месяцев назад

      @@ianbraganza3065 6000 tokens per minute. A token is about 3 characters of text, so around 18k characters per minute

  • @LegionLeague
    @LegionLeague 8 месяцев назад +3

    This seems to be a real game changer for developers, but something feels off. What are caveats of running groq + llama 3 for free? What are limitations? Why is everyone and their grandmother not creating apps with this exact set up right now?

    • @aiunleashed509
      @aiunleashed509  8 месяцев назад +12

      Their is rate limits (I will put them below), so to use it with any volume you would need an enterprise plan. But I still think this is a game changer not just from what groq is doing but in that you now have a top tier model that is open sourced. This allows us to setup our own high performance llm without having to always go through a few megacap tech companies.
      the limits for llama3-70b-8192 are 30 requests per minute & 6,000 tokens per minute. Also max 14,400 requests per day.

  • @catalinabarriosgarciahuido5539
    @catalinabarriosgarciahuido5539 Месяц назад

    When I´m in the terminals while adding the "pip install groq" it says
    pip : The term 'pip' is not recognized as the name of a cmdlet, function, script file, or operable program. Check the spelling
    of the name, or if a path was included, verify that the path is correct and try again.
    At line:1 char:1
    + pip install groq
    + ~~~
    + CategoryInfo : ObjectNotFound: (pip:String) [], CommandNotFoundException
    + FullyQualifiedErrorId : CommandNotFoundException

    • @SuryaPrakash-oe1fz
      @SuryaPrakash-oe1fz Месяц назад +1

      This error shows up on windows when one tries to use pip in the command prompt. To solve this error on windows, you must declare path variable by following these steps:
      -Right click on My Computer or This PC
      -Click on Properties
      -Click on Advanced System Settings
      You will find a section called system variables. Click on Path from the list of variable and values that shows up there.
      After clicking on path click edit. You will find a New button in the pop up.
      Click that and paste the location of the python35 or python36 folder (The location you specified while installing python) followed by \Scripts there.
      For me its C:\Users\suee\AppData\Local\Programs\Python\Python35-32 so I type C:\Users\suee\AppData\Local\Programs\Python\Python35-32\Scripts
      Click Ok to close all windows and restart your command prompt.
      I repeat - restart your command prompt.
      Everything should now be working fine! Make sure you don’t disturb anything else in the path variable and follow the aforementioned steps exactly.

  • @pacific101-t8c
    @pacific101-t8c 4 месяца назад

    Hi , is it imp to download the model and do the setup or just calling api in backend is enough to get ai response does meta provide the api service since i dont know python I rely on api only 💀

    • @aiunleashed509
      @aiunleashed509  4 месяца назад

      No you don't need to download the model. In this video it's run in the cloud with groq(not meta directly).

  • @Tprakh-iw6qt
    @Tprakh-iw6qt 4 месяца назад +1

    Thanks a lot. It was truly helpful

  • @PalkiaOwnsYou
    @PalkiaOwnsYou 6 месяцев назад +1

    Hi,
    Is there a way for the code from this video to be posted?

    • @aiunleashed509
      @aiunleashed509  6 месяцев назад

      Yes, here it is: github.com/AI-Unleashed/GroqResume

  • @tahreersquare
    @tahreersquare 2 месяца назад

    Can I use this free API for n8n without coding?
    I'm a newbie 😊

    • @aiunleashed509
      @aiunleashed509  2 месяца назад +1

      Hi! haven't used it myself but it looks like they have a Groq Chat Model integration: n8n.io/integrations/groq-chat-model/

  • @ashleylauren4133
    @ashleylauren4133 4 месяца назад

    Thanks! Super helpful and straight to the point

  • @JackFlas
    @JackFlas 7 месяцев назад

    Great vid! I could manage to make it work, but not with all the code you gave. Either it writes an empty .md or else I erase some of the code and gives me the cv on the output console but it doesn´t create the .md file.
    This is the error I get: "line 36, in
    output_file.write(completion.choises[0].message.content)
    AttributeError: 'Stream' object has no attribute 'choises'"
    Could you give me some hints?

    • @aiunleashed509
      @aiunleashed509  7 месяцев назад

      Hey Thanks for watching. I will post a link to the GitHub with the starter code shortly. Noticed you have a typo in the word choices, should be: completion.choices[0].message.content

  • @shivamjoshi6648
    @shivamjoshi6648 3 месяца назад

    Thank You so much sir for providing such rare information , I can now use powerful ml models for free

  • @MedhanshSharma
    @MedhanshSharma 4 месяца назад

    i want to use it in spring boot project how to do that

    • @aiunleashed509
      @aiunleashed509  3 месяца назад +1

      At a high level you want to implement a service class that will handle the API calls(the code from this video would go in that class). Then use the service in your controllers or other components. ChatGPT or Claude could walk you through it in some more detail, good luck!

  • @lifewithG-bengs
    @lifewithG-bengs 4 месяца назад +1

    Wait what so this is completely free?

  • @KoratNewsTH
    @KoratNewsTH 5 месяцев назад

    so amazing! Thank you so much

  • @godsfavourofuyovwi
    @godsfavourofuyovwi 12 часов назад

    I don’t think this is free, just checked they charge for tokens

  • @vigneshrs3221
    @vigneshrs3221 Месяц назад +1

    Is it still free?

  • @mauricemaeterlinck8799
    @mauricemaeterlinck8799 5 месяцев назад

    Amazing. Thank you. got my sub :)

  • @MengLu-k6i
    @MengLu-k6i 2 месяца назад +1

    Super super good!!!!

  • @SANCHEZREGIEMARKB
    @SANCHEZREGIEMARKB 2 месяца назад

    Is it free? and No Limits?!

    • @aiunleashed509
      @aiunleashed509  2 месяца назад +1

      yes its free, no credit card required even. It does have some request limits but they are pretty high last time I checked. I am a bit surprised it is still totally free with such high usage limits...

    • @SANCHEZREGIEMARKB
      @SANCHEZREGIEMARKB 2 месяца назад

      @@aiunleashed509 Wow, I'll try this, BTW Thank you.

  • @_angThanhLoc
    @_angThanhLoc Месяц назад

    so usefull, i like it

  • @yulukagigantemuriel4181
    @yulukagigantemuriel4181 6 месяцев назад +1

    You're awesome!

    • @aiunleashed509
      @aiunleashed509  6 месяцев назад +1

      Thanks, comments like this keep me going. You're awesome too!

  • @TimeIsHQ
    @TimeIsHQ Месяц назад

    THE GOAT💎💎

  • @Yuri-wk8rk
    @Yuri-wk8rk 7 месяцев назад

    Thank you, please post source code next time ! :)) I havу tested it, works perfect

  • @all_in_java9347
    @all_in_java9347 2 месяца назад

    thanks

  • @renify_
    @renify_ 25 дней назад +1

    Its free as of the moment LOL

  • @playgirl44
    @playgirl44 5 месяцев назад

    when i do the 'python project.py' step in terminal it doesnt work. how to fix this?