STOP PAYING for Github's Copilot with this LOCAL & Free Code Assistant

Hey👋 if You Enjoyed This Video and are interested in more content about AI, Agents and Business Automation -- please Subscribe🙂
---------
In this video I'm sharing a local and Free coding assistant which you can use with Ollama (if you have GPU) or using Groq API.
🎞️Chapters
00:00 - STOP PAYING for Github's Copilot with this LOCAL & Free Code Assistant
▶️Other Videos You Might Like
Autogen Just Got Better - Controlling The Agents With Validation and State Transitions:
• Autogen Just Got Bette...
Self Reflecting Agents - Autogen Producing Amazing Production Level Results (Real Business Use Case)
• Self Reflecting Agents...
Autogen Agents with RAG - Upgrading Agents with a Knowledge Base
• Autogen Agents with RA...
💡Mentioned in the Video
evalplus.githu...
huggingface.co...
www.continue.dev/
console.groq.c...
🌐Links
✅Come say 'Hi' over Linkedin:
/ yaronbeen
✅Book a Consultation:
ecomxf.com/boo...
✅My Website:
ecomxf.com/
#️#️#️
#codingassistant
#ContinueDev
#GithubCopilotAlternative
#ollama
#groq
#LocalCoding
#programminghelp
#codegeneration
#pairprogramming
#AIforCode
#DeveloperTools

[@mazfrost]

Does this have access to all files and more importantly use them contextually with each other? I'm looking for an AI tool that can view an entire JS project and compare different files, and edit based on them...
Currently when working with ChatGPT or Opus they do a pretty good job at coding, but because they don't have entire project context, they will make "renaming" mistakes. So they will rename a Base URL a little bit or make up a route that doesn't exist, and small errors like this are hard to find when dealing with a huge project, and it makes the debugging process take a while, especially when the errors are sneaky or just slight mistakes. I've tried OpenDevin/Devika, but they aren't quite there yet, or maybe I'm just not setting them up right. Thoughts? Apologies for such a long comment.

[@YaronBeen]

It is supposed to be able to do this.
OpenDevin and Devika are exciting projects with potential, but at the moment they are not really useful (unless your goal is to get UA-cam views).
I don't think you didn't configure them correctly,
They just don't work so well yet haha

[@wiliamferraciolli5380]

hi, ta for the video. can Llama be run on docker rather than installing locally?

[@YaronBeen]

Haven't tried it.
But you can run Ollama on Docker
hub.docker.com/r/ollama/ollama

[@YaronBeen]

Check this video if you want to see how I built an MVP in less than 5 minutes:ua-cam.com/video/ETJwarM7x3Y/v-deo.html

[@PriNovaFX]

Thank you for this video. If you click on the tags link in the Ollama model library you can compare the sha-checksum to see that chat, latest and 7B are the same. So, no need to switch to :chat model.

[@YaronBeen]

Thanks for noting this :)

[@SA-lh6bx]

Concise and great video as usual.. This is time and cost saver.. Thanks...

[@YaronBeen]

:)
Glad you find it useful

[@DalazG]

I'm looking for some cold helper that is able to observe all my files in a project. Can this do that?

[@YaronBeen]

docs.continue.dev/how-to-use-continue#ask-questions-about-your-codebase

[@DalazG]

@@YaronBeen thanks!

[@YaronBeen]

:)

[@Dreamslol]

To be fair, Continue is a copy of Cursor which is way better imo

[@JBLU7]

why is it way better? what differences?

[@GuillaumeMOREL-jb5rr]

Thanks

[@YaronBeen]

Sure thing!