FREE: Ollama Engineer (Llama 3.1) Claude Engineer Alternative! 🤖🦙 AI Coding Assistant Open-Source👨‍💻

What's going on everybody Josh pook here and in today's video we're going to be looking at how you can use oama with not claw engineer but the makers of Claude engineer they have a new tool called olama engineer all right we're going to show you you can set it up self host it well it's something that you just it's a pip Python program so you just set it up run it and then it acts sort of like CLA engineer a lot

Of people on my CLA videos whether it was CLA engineer CLA developer whatever the case may be we're saying hey how can we do local models or AMA you know quad's a bit pricey so in this video we're going to show you exactly how to do that let's dive right into [Music] it all right guys so if you're not familiar with CLA engineer um check out my other video on that it's actually right here so this is by the same creator

Of Claude engineer and um yeah so they came I mentioned in that video if you want me to do a video on uh AMA engineer let me know so here it is and basically I'm not going to go over all the intricacies of Claude engineer and the whole repo again but this is the repo link for it will be in the description down below you can see here it's sort of like if you're familiar with AI pair programmers maybe like AER

Or uh Claude developer it's kind of like that Claud Dev um where it's a CLI um code companion it's not like G well it's not like um continue Dev or get help copilot within VSS code or a vs code extension it's within the CLI so similar to AER or quad Dev there's other ones as well um okay so basically uh it's actually I guess it's I mean it's pretty easy to get a set up first things first you're going to want

To get oama if you don't have oama already go to oama link will be in the description you can download it okay now with ol llama engine or llama engine you can use it with any specific model open source model you want whether it's 53 mistro DEC coder V2 uh Gemma or llama 3.1 just cuz it's the newest one we'll use llama 3.1 so you just go here you select your parameter we're going to use the 8B model which you're going

To use if you're using llama 3.1 it's the one that's the least Labor uh resource intensive so you're just going to copy this command AMA run llama 3.1 and you're going to run that in your uh in your uh command line once you actually install o llama all right once you do that um you're going to be able to go back to the GitHub repo and you can either go up here and download the zip and then place it on wherever

You want whether your desktop or whatever um or you can just get clone and use get and clone the repo and run that in your command line you would just copy the link all right I already have it I cloned so I won't be doing that once you do that you're going to change directory in your command line to wherever clad clad engineer is cloned in okay and then what you're going to do is you can do pip install dasr requirements

So it's the similar process to how claw Engineers basically installed so it's actually it's basically the exact same process so you get clone the repo you CD into CLA engineer then you pip install R Dash or- R requirements.txt okay then you're going to set up your environment variables so you could set this up as like um setting these as variables and if you're on Windows or exporting them if you're on Linux or Mac but you I just set them up within

The EnV file so I believe there is a example template right here EnV uh. example so I just changed this to EnV and then I used um I pasted in my quad GPI key as well my anthropic as well as my tavali okay because you're going to use tavali it basically is to search the web so this is tavali right here pretty cool it's a pretty useful tool and it's made specifically for uh AI search engine optimize or searching the with

Llms so they give you I believe 100 or a th000 free Searchers something like this a decent amount maybe a thousand and uh I still haven't ran through all my tavali free on so get that you can just sign up for free put your API key here now I will say this this is something that I mentioned in my Claud um engineer video and this is something I'll mention in this llama video as well right you're going to run into issues

If you just try to install the requirements and then put the API keys in and then you try to run um llama engine all right so first off I'll say if you want to run clae engine then you just basically do it like this right here you can uh you could create a virtual environment and then activate it and install the requirements and then run python uh main.py so Claud engineer is the main.py file and what you'll see here I have

It open in VSS code main.py and then AMA engineer is the olama engineer. pi so depending on what you want to use you could run either one of those but you will encounter issues if you don't do this step I'm going to show you basically as of now with the recent um Tav update basically as of now I mean at the time of recording this maybe in the future it won't be like this but it basically has a few dependencies and

Whatnot that require you to put a cohere API key which we don't want to do because why would we want to do that it's just like an extra hassle thing so what you can do is you can simply just go Um so let's just see here yeah you'll see uh where is it basically you're going to do pip uninstall uh tavali just search um so pip uninstall Tav yeah right here pip uninstall tavali . python so you're going to uninstall this

All right and then you're just going to say would you like to remove you're going to say yes and then once it's removed you are going to go um you're basically going to run this command and I'll leave this command in the the description you're going to run pip install Tav do- Python and then equals equals 0.3 uh five all right this is the version um that is the was before they made these recent updates so you're going to want that

Version okay so it's not going to require you to put a coher API key in okay once you do that um you should just to ver just to confirm everything you'll have your enir environment variable set up you'll have your requirements set up you'll have the proper tavali set up and then at that point you can simply just run python ol- engine. now actually before you do that I want to say one more thing if you you open up your oama

Engine python uh file or if you just open up the whole CLA engineer folder and VSS code and you go here you're going to want to scroll down uh right about here line 78 and these right here 81 82 you'll see here I have llama 3.1 initially it was I believe uh mistol or something like this you can put whatever Umama model you want here just make sure you obviously have this downloaded and you're referencing the right name okay and then

Also too if you're going to want to use autoc complete mode you can change the max continuous iterations right here by default it's set to 25 and then the max to context tokens is set to by default 200k so you can actually look through um some of these some of these uh this code right here and potentially alter some of it most it's pretty short it's not too okay I guess it's somewhat long but yeah this is some of the tools

I believe I haven't read through a lot of it but yeah if there is any other those are the main customizations you're going to want to do changing the model Etc and then from that point you can run it there may be a few other things if you want to check in there and customize it a bit to your liking then you can simply just run this Command right here python - engine. and it's going to say welcome to AMA llama

3.1 engineer chat with multi-agent and image support so that's one cool thing they have image support um so you can exit auto mode reset save and chat so while in auto mode plus control C at any time to exit auto mode this is something like I've had mixed reviews on like mixed success on in the past I mean I haven't like built anything crazy using auto mode in the future yes obviously like I think it's going to be really cool when

AI can like VAR well iterate upon itself with the code I just personally don't think it's like at that um that like that top TI your level yet um I mean it really isn't um but it is kind of fun sometimes so let's go auto mode and so let's just say build me a um a pong pong game in python keep track of score and have me play a computer I don't know how well it will be able to do this

Maybe a little bit complex I don't know or to play the computer but we'll come back when it's done all right guys so I'm back after trying to run some of this now I'll let be completely honest with you guys I was trying to run that First Command with the auto uh running uh auto mode and I was having some issues with auto mode maybe it's just me I don't know I put it down to 10er durations but still was having

Some issues um with just a few different things not creating the file not creating a proper game maybe it's a little bit too complex a pawn game I don't know um or maybe it's something I'm doing wrong me personally just after using it I would probably not use this on a day-to-day basis for you know actually developing or coding or building anything I would maybe use it here and they are just kind of for small little things maybe um but at

The same time it really makes no sense for me to do that just because I have tools like ader I have tools like continue Dev I have tools even like cursor uh even um quad Dev even CL honestly quite frankly even Claude engineer I was getting uh but obviously that's you know you pay for API but you can use AMA with things like continued Dev with AER Etc so um with this I found it just to be frankly fairly slow um

I mean it is local models so I guess but uh and I was having issues um actually even creating the game so I actually it did create some code and I for me at least it wasn't working properly so I actually use a different AI model to pretty much just like try to fix it and as you can see now it's like a working game it's pretty decent in terms of like I mean I'm not playing as a computer but it's

Pretty decent here I mean it's just going back and forth but this is not that is not 100% the doing of this tool so I can't give it credit to it um I simply just was like Hey creman HTML page just saying subscribe it manage to do that which is I mean if you want to do something like that or small things here and there maybe trying different models out on a llama you get get it to work better maybe I'm

Missing something here but other than that guys just want to cover this because this is um you know CLA engineer a lot of people know about Cloud engineer this is llama engineer so will I you be continuing it no but if you do want to test it out you can follow these steps along see for yourself because at the end of the day this is my opinion maybe you'll actually like it maybe I'm missing something so you can give it a

Try for yourself let me know in the comments Down Below guys and let me know what your favorite tool is whether it's AER continu div uh cursor whatever the case may be if you want me to cover any other ones let me know in the comments down below and if you're new here we upload videos every single day on AI automation business growth Marketing sales Etc so make sure to smash that like button smash that subscribe button if you got some

Value here other than that I will see you in tomorrow's video guys keep hustling keep grinding and of course guys accelerate your stride take care