Want to HIRE us to implement AI into your Business or Workflow? Fill out this work form: td730kenue7.typeform.com/to/WndMD5l7 💗 Thank you so much for watching guys! I would highly appreciate it if you subscribe (turn on notifcation bell), like, and comment what else you want to see! 📆 Book a 1-On-1 Consulting Call WIth Me: calendly.com/worldzofai/ai-consulting-call-1 🔥 Become a Patron (Private Discord): patreon.com/WorldofAi 🧠 Follow me on Twitter: twitter.com/intheworldofai Love y'all and have an amazing day fellas. Thank you so much guys! Love yall!
Mind-blown by Llama 3.3's capabilities and cost-effectiveness! Love how Llama Coder simplifies generating web apps, definitely going to try it out . Your breakdown of the features and examples really helps, thanks for sharing!
As an opensource solution, its the best thats out there (as in using an opensource framework + opensource llm). However, I still believe claude 3.5 sonnet is the best at coding.
@@intheworldofai the new gpt o1 pro would beat all the coding llms by a significant margin.. once it is available through API. Can't wait to see people integrating and building better and better agents
Well, when i'm on npm install the cmd says " 'npm' is not recognized as an internal or external command, operable program or batch file. " and i cant go furder .
Totally agree about not have a folder tree to view files, or an integrated terminal option. Maybe someone will port llamacoder over to vscode why reinvent the wheel.
When I try to publish I get an error in the bottom left corner. When running on Windows, file names are checked against an access list to protect against remote code execution attacks. File names may consist only of alphanumeric characters (all languages), periods, dashes, slashes, and underscores.
I'm an ai developer 😂I've built multiple apps over the past 2 years since the launch of open ai which I started, I moved to claude sonnet then gemini I have to say gemini has improved and in my opinion is better at holding the actual context of a bigger code base, which it doesn't assume code functionality like claude sonnet does often breaking my code here or there so with that gemini is better than claude on bigger code bases but sonnet is fire asf to but just gotta be more cautious when making bigger updates
I find the same with Sonnet. I found Gemini the best for image recognition. Will def try Gemini for coding. How are you using Gemini with coding tasks? I use Sonnet with Cursor mostly - which communicates well with codebases.
@@antigravityinc so basically I just prompt it with what I need but claude is best for building the initial app structure and getting the app setup then I use gemini when the app becomes more comprehensive and complex.. but I just copy and paste but at same time check their work I don't use any 3rd party apps like cursor, aider they tend to break code especially on bigger code bases. I'm actually building a course on how I excel at developing using ai
I dont understand the practical use for this can you explain it. It seems it can just generate some good baseline code to get you started but there isn't really a way (at least from what I can see) to chat with the model about implementation details and changes. For example, if the code it gives me requires debugging I can't discuss what might be the issue with the model. Am I misunderstanding this or can it do that?
[Must Watch]: Llama-3.3: The BEST Opensource LLM EVER! Beats GPT-4o! (Fully Tested): th-cam.com/video/MKYAiCDySnE/w-d-xo.html Run ALL Your AI Locally in Minutes! LLMs, AI Agents, RAG, and More! (Opensource): th-cam.com/video/MqzEnwP0s60/w-d-xo.htmlsi=SJsoH_UJXr5Sg6DO Phidata: NEW Agent UI - Build Agents with Memory, Knowledge, Tools & Reasoning! (Opensource): th-cam.com/video/QGeecf3wteI/w-d-xo.htmlsi=s1go2X8K8dxxnShD
Lmao I don’t think so. Other models will also get better. Plus, AI intelligence growth has slowed down. We don’t have models coming out significantly smarter anymore. Even huge params increase doesn’t really make them much more intelligent
@@mr_ribin_123 Oh sorry haha. Unfortunately you can't download the code :( I will recommend this feature to the dev. Best thing to do is copy and paste the code snippet.
Terrible. Only works for simple applications. I did the frontend, backend, software integrations just using cline dev. Claude himself is times better, nothing compares so far.
Want to HIRE us to implement AI into your Business or Workflow? Fill out this work form: td730kenue7.typeform.com/to/WndMD5l7
💗 Thank you so much for watching guys! I would highly appreciate it if you subscribe (turn on notifcation bell), like, and comment what else you want to see!
📆 Book a 1-On-1 Consulting Call WIth Me: calendly.com/worldzofai/ai-consulting-call-1
🔥 Become a Patron (Private Discord): patreon.com/WorldofAi
🧠 Follow me on Twitter: twitter.com/intheworldofai
Love y'all and have an amazing day fellas. Thank you so much guys! Love yall!
I dont think this is yet ready for the coding. Claude 3.5 sonnet still beats everything
Claude 3.5 is truly amazing but this does get the job done.
This is a free model. We need to consider that too.
Claude 3.5 is amazing for everything...
How we configure Claude 3.5 with clide?
Can Claude run locally?
Mind-blown by Llama 3.3's capabilities and cost-effectiveness! Love how Llama Coder simplifies generating web apps, definitely going to try it out . Your breakdown of the features and examples really helps, thanks for sharing!
Appreciate it, glad you enjoyed the video!
Awesome. Really needed that for coding an enterprise grade RAG
As an opensource solution, its the best thats out there (as in using an opensource framework + opensource llm).
However, I still believe claude 3.5 sonnet is the best at coding.
@@intheworldofai the new gpt o1 pro would beat all the coding llms by a significant margin.. once it is available through API. Can't wait to see people integrating and building better and better agents
Llama 3.3 sounds like a game-changer! Thanks for sharing your experience with Llama Coder, I'm definitely going to check it out. Great video!
Well, when i'm on npm install the cmd says " 'npm' is not recognized as an internal or external command,
operable program or batch file. " and i cant go furder .
Totally agree about not have a folder tree to view files, or an integrated terminal option. Maybe someone will port llamacoder over to vscode why reinvent the wheel.
Yeah, honestly thats the best thing to happen for this app for future development. Similar to how it was with the bolt.new fork
When I try to publish I get an error in the bottom left corner.
When running on Windows, file names are checked against an access list to protect against remote code execution attacks. File names may consist only of alphanumeric characters (all languages), periods, dashes, slashes, and underscores.
good update! thanks for your sharing
Thanks for watching!
I'm an ai developer 😂I've built multiple apps over the past 2 years since the launch of open ai which I started, I moved to claude sonnet then gemini I have to say gemini has improved and in my opinion is better at holding the actual context of a bigger code base, which it doesn't assume code functionality like claude sonnet does often breaking my code here or there so with that gemini is better than claude on bigger code bases but sonnet is fire asf to but just gotta be more cautious when making bigger updates
I find the same with Sonnet. I found Gemini the best for image recognition. Will def try Gemini for coding. How are you using Gemini with coding tasks? I use Sonnet with Cursor mostly - which communicates well with codebases.
@@antigravityinc so basically I just prompt it with what I need but claude is best for building the initial app structure and getting the app setup then I use gemini when the app becomes more comprehensive and complex.. but I just copy and paste but at same time check their work I don't use any 3rd party apps like cursor, aider they tend to break code especially on bigger code bases. I'm actually building a course on how I excel at developing using ai
@@AntonioMcGee-hk1kqGemini is horrible if you want something complex, but if you give it a code to debug, it can outperform everything
Few questions
What are the requirements for running it locally?
What is the download size?
How does it compare to qwen 2.5 coder 32b?
1. If you are using the togetherai's api, its quite light weight. Most computers can run it.
2. Around 200-350 MB
3. Haven't tested yet
@intheworldofai well, I meant running the LLM locally
@@amit4rou bro how to get the api?
@@Bhuvan.Shetty it's paid, that's why I want to run it locally
From where am I supposed to get api key from?
I tried it but it sucks. It produces errors the whole time for a simple website I want to build.
'npm' is not recognized as an internal or external command,
operable program or batch file.
what should i do with this?
You need to install NPM package. Just paste in that command to chatgpt. It will resolve the issue for you by providing the correct command.
So i can use it to develop an app that's a wrap around google maps no code
Yes, if you prompt it correctly.
Can I run it with ollama? Can I give a sample code?
I dont understand the practical use for this can you explain it. It seems it can just generate some good baseline code to get you started but there isn't really a way (at least from what I can see) to chat with the model about implementation details and changes. For example, if the code it gives me requires debugging I can't discuss what might be the issue with the model.
Am I misunderstanding this or can it do that?
you can paste in the error message and ask it for the fix, or use some current available IDE that let you hook up your own model for AI assistance
Can we trust Meta though?
Only React, right?
Fully local but you need a high end graphic card. My 12GB GPU was no match for it.
Bolt.New UPDATE: Develop a Full-Stack App Without Writing ANY Code! (Auth, Database, Etc!): th-cam.com/video/4DnAHDk9Iyc/w-d-xo.html
[Must Watch]:
Llama-3.3: The BEST Opensource LLM EVER! Beats GPT-4o! (Fully Tested): th-cam.com/video/MKYAiCDySnE/w-d-xo.html
Run ALL Your AI Locally in Minutes! LLMs, AI Agents, RAG, and More! (Opensource): th-cam.com/video/MqzEnwP0s60/w-d-xo.htmlsi=SJsoH_UJXr5Sg6DO
Phidata: NEW Agent UI - Build Agents with Memory, Knowledge, Tools & Reasoning! (Opensource): th-cam.com/video/QGeecf3wteI/w-d-xo.htmlsi=s1go2X8K8dxxnShD
Llama 4.0 will be out in 2025. It will be even better. It can conquer the world too. Wow!!!
Lmao I don’t think so. Other models will also get better. Plus, AI intelligence growth has slowed down. We don’t have models coming out significantly smarter anymore. Even huge params increase doesn’t really make them much more intelligent
Agent-Zero UPDATE: Autonomous AI Agent Framework Can DO ANYTHING! (Generate Full-Stack Apps): th-cam.com/video/mm4g_BaNZu0/w-d-xo.html
Gemini 2.0 Flash + Cline: Develop a Full-stack App For FREE Without Writing ANY Code!: th-cam.com/video/IDdYU8IKglk/w-d-xo.html
But how to download the code?
Clone the repo
@intheworldofai No, I mean After creating the app😊 ít has only publish function to see but what for downloading and using that code?
Just copy paste?
@@mr_ribin_123 Oh sorry haha. Unfortunately you can't download the code :( I will recommend this feature to the dev.
Best thing to do is copy and paste the code snippet.
@@intheworldofai 👍🏻 okay ... sure
is this unlimited and free?
Yup!
Claude 3.5 sonnet still demolishes it and also Cline is much better than this app
100% I agree but this app is quite impressive and its a great opensource model
Warp: AI Terminal - FREE AI Coding Agent BEATS Cursor, V0 & Bolt! (Generate Full-Stack Apps!): th-cam.com/video/uRquE0FjvFk/w-d-xo.html
Terrible. Only works for simple applications. I did the frontend, backend, software integrations just using cline dev. Claude himself is times better, nothing compares so far.
You vídeo is genial . Plis conecte multilenguaje traducción you videos
Thank you. Will try to
Hope this API is permanently free but maybe that’s not likely
Looks like a stripped down Open-WebUI