OttoDev REVOLUTIONIZES Bolt.new and AI Models in 24 Hours!

[@johngoad]

Glad to meet you, great video, I have been playing with it for a couple weeks now. I've have barely built anything with it, just built on it

[@SSmartyimages]

You do have a good voice for listening:)

[@vincibits]

Thank you :)

[@ChetanJariwala]

Thanks for putting up this useful video. Another fan of your calming voice 😊. I am not a developer but know enough to create basic web stuff. I would really appreciate it if you can make a video on how to actually take the code and create an app or website from bolt to your own custom domain or cms. I haven't seen anyone create a video on this particular topic.

[@vincibits]

Great idea! Will definitely do that!!

[@youssefjbida7038]

Yed that would be a great idea and a generous gift to we beginners. Because what we need is an A to Z tuto. No one did it yet. Everybody talk about bolt.new and AI but no one goes until publishing the apps and make it public on play store or apple store. Thanks in advance

[@Ph1lH4ck]

I'm simple guy ! When I see Vincibits ! I click the like button 😉

[@vincibits]

You’re too kind!!

[@huntinggamer1725]

Love your channel joking the community soon

[@vincibits]

I appreciate that! Thank you so much for you support! It means a lot. Thank you!

[@svenvarg6913]

Crucial bit for me was on the preview not working, I spent 2 hours going through videos, docs wondering if I missed something and was just about to do dig into the code to debug the hell out of it. The preview bug just fails silently So a big thank you 😎

[@vincibits]

I’m glad this helped :)

[@Ph1lH4ck]

Let's built an ElevenLabs Vincibits voice model to spread the Cool all over KZbin 😎

[@vincibits]

Haha yes!! And then the world would just not care lol

[@TeemuKetola]

What kind of PC specs do I need to run Gwen 2.5 32B quantized version locally with Bolt? Will RTX4090 and 128GB RAM be enough?

[@vincibits]

I would just give it a try and see :)

[@MatheusSilva-qm3ph]

Very good bro! One question is what is the minimum PC configuration to run locally? I did not find this statement. Thanks

[@vincibits]

It all depends on the model, but at minimum 10GB or RAM and fairly modern PC would do it, maybe not the best but it would do.

[@AssassinUK]

Do you know if it can connect to Openrouter? You can then use that to run the models, including the big ones.

[@vincibits]

Yes you can, look at the .env file there’s a place to add the api key

[@Ph1lH4ck]

The computer 👉 🥵😫🤯 😂

[@vincibits]

Oh it was bad!!!

[@wolfgangdelaine5284]

Thx for your videos ! And for ollama installation ? I installed with docker, and bolt too. Ollama alone take 3 times less ram than with bolt. You know why ?

[@vincibits]

I’m not sure, but I’d suspect they there just a lot of moving parts as a system itself. Remember, you’re not just running illama when you use Ottodev - it’s an application that has many other dependencies etc.

[@myWorldDiscover]

good explanation. thanks 🎉. Can you please build a complex solution?

[@vincibits]

Yes of course, what did you have in mind?

[@myWorldDiscover]

@@vincibits like ERP System, or Discord, or CRM app, i mean a real complex and complete web app. not just todo, calculators. these things are so simple. please don't forget both backend and frontend. thanks in advance

[@saidabedi2836]

Getting really hard to keep up with the constant updates in the AI world. Any advice keeping up?

[@vincibits]

I know. I hear you. Not trying to push anything on you, but I actually started a community group that mitigates this same problem of overwhelming feeling when it comes to AI. Let me know if you are interested in learning more about it.

[@saidabedi2836]

@@vincibits interested in the community. Wanted to know what courses are offered.

[@vincibits]

@@saidabedi2836 Hi, the community is not just about courses (we do have courses), but it's about having access to me and other highly motivated and like-minded people who want to (learn) and build AI solutions that solve real-world problems. We have weekly/monthly calls to review project workflows, use cases, Q&A, etc. We also have targeted tutorials (courses) ranging from mastering Python, building your first AI Chatbot, and learning how to build full-fledged AI solutions. It's a growing community, and we are always looking at what's new in the AI world and continue to evolve that way. Most importantly, you'll have access to other members and me to discuss issues you might have etc... Does this make sense? Let me know if you have any questions at all. You can join now, you won't be disappointed: www.skool.com/vincibits/about

[@SpkSpkS]

I have a fine tuned llm and how to use it ?

[@vincibits]

You would use it like you would any LLM.

[@JOJO-ck7cm]

Can someone build an interface onto OttoDev to prevent all this problems

[@vincibits]

Good question. This is why a lot of people are contributing towards the project on Github. It’s a work in progress type of project.

[@Ph1lH4ck]

By the way I have installed Bolt with Docker and connected it with Claude, OpenAI and LLama to do some tests. And everything works fine. But the output of apps built locally are way less good and powerful than the online bolt.new app even with the Claude Sonnet API. So I'm a bit disappointed by the local solution.

[@vincibits]

I have noticed somewhat similar symptoms as well… I think there’s something else that Bolt.new, the original one, has left out from the open source repo. We’ll see.

[@Ph1lH4ck]

@@vincibits yes definitely

[@Ph1lH4ck]

@@vincibits I've tested windsurf in my computer today and I like it better than bolt because you learn the tech by using it. And it's working 🙂

[@vincibits]

@@Ph1lH4ck Interesting! I will check that out as well. Thanks!

[@learn_generative_ai]

Hello I don't know why but I can't able to pull my api key from .env.local but when I gave explicitly in the execution time it's working what do think the error was?

[@vincibits]

First you have to change the .env.local to just .env

[@learn_generative_ai]

@vincibits as recommended in the git repo readme I have changed it from. .env.example to .env.local but that doesn't worked also I changed it to just .env still it doesn't works

[@PauloDichone]

@@learn_generative_aithen I think you have something else that is not right. Are you running docker application? It’s hard to troubleshoot not knowing what you have setup

[@learn_generative_ai]

@PauloDichone Yes, It was running on docker