I Replaced GitHub Copilot with Gemma 4 (Zero Cost, 100% Offline)
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Related videos
I Replaced 16:8 Fasting With This and Everything Got Better
Thomas DeLauer
44.1k views
Spending $40,000 in Copilot for $40
Theo - t3․gg
84.6k views
Get In, We're Leaving GitHub
Theo - t3․gg
137.1k views
I Replaced 10 Power Tools With My Drill!
Project Farm
277.4k views
Uber CEO: I Have To Be Honest, AI Will Replace 9.4 Million Jobs At Uber!
The Diary Of A CEO
72.8k views
LOCKED into a conflict with NO reverse gear
The Duran
38.0k views
The Science of Replaceable Parts, with Mary Roach
StarTalk
39.9k views
$10M CEO: How to Get Ahead while Others Get Replaced | Daniel Priestley
Silicon Valley Girl
107.7k views
Replacing Humans with AI is Going Horribly Wrong
ColdFusion
1.4m views
So I tried the GitHub "Vibe Coder"...
Theo - t3․gg
60.6k views
Top Comments (10)
Its like comparison between a PHD holder with a 10 year old 😂
Not for production level code. At best, you can create a site and you’ll be lucky if it semi-works. Good luck on iterating on it. 3 days to integrate a new button that calls an api. Still is trying to work with the tooling looping. Sending only part of the necessary schema. They need to improve the model on their tooling calling mechanism.
Thanks for this bro, as a tech enthusiast with no CS background, I am learning about Git thanks to your videos! If it's a complex project and there is a hardware limitation - we can borrow hardware (neocloud for example) and use local models if privacy is a concern. I think it's brilliant that we have so many options to choose from to optimize according to the project and use case.
Can you upload a video showing how to integrate a local LLM model with the Spring Boot Embabel agentic framework?
Great video Sir
17:26 setting context window to max will consume lot of memory
Excellent video, thank you for sharing this setup. I wanted to ask: how can we use the GPU similarly to an NPU for local AI inference? Is there any specific configuration, model optimization, or backend required to fully utilize the GPU for faster offline performance? Would appreciate it if you could explain this part as well.
Nice job Dude
I already have a mid tier gaming machine, I can try this 🤞
So cool
Unlock the Data Inside
Turn Videos into Knowledge
- Get FREE 10/day: transcripts, summaries, chats
- Chat with videos, export text & PDF
- $1 free API credit for RAG, chatbots & research
Free forever plan • All features unlocked
Top Comments (10)
Its like comparison between a PHD holder with a 10 year old 😂
Not for production level code. At best, you can create a site and you’ll be lucky if it semi-works. Good luck on iterating on it. 3 days to integrate a new button that calls an api. Still is trying to work with the tooling looping. Sending only part of the necessary schema. They need to improve the model on their tooling calling mechanism.
Thanks for this bro, as a tech enthusiast with no CS background, I am learning about Git thanks to your videos! If it's a complex project and there is a hardware limitation - we can borrow hardware (neocloud for example) and use local models if privacy is a concern. I think it's brilliant that we have so many options to choose from to optimize according to the project and use case.
Can you upload a video showing how to integrate a local LLM model with the Spring Boot Embabel agentic framework?
Great video Sir
17:26 setting context window to max will consume lot of memory
Excellent video, thank you for sharing this setup. I wanted to ask: how can we use the GPU similarly to an NPU for local AI inference? Is there any specific configuration, model optimization, or backend required to fully utilize the GPU for faster offline performance? Would appreciate it if you could explain this part as well.
Nice job Dude
I already have a mid tier gaming machine, I can try this 🤞
So cool