All the Coding Assistants 🤯 All at once!
HTML-код
- Опубликовано: 19 авг 2024
- This is the assistant I've been using: thinkbuddy.ai/ Use the code ALEX to get a nice discount
👀 My main machine: amzn.to/4cfBBPH
👀 M3 MacBook Pro SALE: aamzn.to/3PZn8yF
⚡M3 MacBook Air: amzn.to/3xdD1e0
👀 M2 MacBook Air Deal: amzn.to/49mON3r
👀 My favorite external drive (dependable): amzn.to/3Os9Wi3
👀 Thunderbolt 4 dock: amzn.to/3yVRicC
⚡ Other gear I use: www.amazon.com...
🤯 M3 MacBook Coding Performance - • 🤯 SURPRISING M3 MacBoo...
▶️ M3 MacBook Air for Developers - • M3 MacBook Air after a...
▶️ INSANE Machine Learning on Neural Engine - • INSANE Machine Learnin...
▶️ Ultimate Web Developer MacBook - • Ultimate Web Developer...
▶️ This is what spending more on a MacBook Pro gets you - • Spend MORE on a MacBoo...
▶️ Apple Silicon and Developers Playlist - • Apple Silicon and Deve...
Developer productivity Playlist - • Developer Productivity
- - - - - - - - -
❤️ SUBSCRIBE TO MY RUclips CHANNEL 📺
Click here to subscribe: / @azisk
- - - - - - - - -
Join this channel to get access to perks:
/ @azisk
- - - - - - - - -
📱LET'S CONNECT ON TWITTER: / digitalix
#macapps #chatgpt #macbook
Use code ALEX for 30% off 🤯 thinkbuddy.ai/
@@AZisk hi alex, i was trying to run mistral large using ollama on my macbook m3 max 36 gb ram but it failed to run it...can you help please...i am new to mac
@@vinprocksss I doubt 36gb will be enough to run mistral large instruct 2407 (if that's what you're talking about).
That's a 123B Model where the smallest quantised version of it (iq 1 S) is about 26gb big.
Add to that the kv cache that gets added depending on the context size + Apple not allowing all of the ram to be used by a single application, you simply might not have enough ram to run that model.
Came to see if my job is at risk
Always :) Keep on learning!
@@AZisk Thank you Alex! Love your content!!
@@AZisk Alex any chance you are considering 60 FPS videos on your channel?
As long as these tools can't do the thinking for you you're good. Especially as a developer.
@@saipulivarthiwhat would be the benefit?
6:25 "this is a keyboard" Bro Gemini I can't lmao
What is the security risk with this company essentially acting as a man-in-the-middle by storing your prompts and data before sending them to a cloud LLM?
I assume you also lose Claude Artifacts when using this service.
No they are storing datas on your local device
All data is stored locally on your computer. You can use Wireshark-like agents to verify that. We only use a server to track usage but never store any messages, directly proxying to the relevant AI endpoint.
@@julsezer651 are you sure about that? If they are paying for the API calls, they wouldn't store their keys in your local machine. It would be too easy to open wireshark and extract the keys. It makes a lot more sense to proxy those API calls to their servers so they can protect their keys. And that makes @julsezer's question more than relevant. Now I will have to install it and check :-)
Of course it's Mac OS only. Disclosure early in the video would be an honest thing to do. "I've been using it for X time on my Mac" doesn't really cut it.
@@Dero02 In their roadmap it states they are working on a version for Windows and iOS.
f nothing useful unless it comes out with a foss client which we can then use on linux.....
@@writer9999 linux 💀 was that the joke of the day?
I mean this youtube channel does do mostly mac stuff, so it is to be expected...
Last few x elite videos and only few minutes in this this video ...
Alex you have become my geektainment ...
Voice commands might be useful when im lazy
looks so promising, definitely will give it a try
Exactly what i was looking for ! Thanks man ! 👌
The remix model allows me to use different languages together. I love this feature.
I remember when search engines were the "new big thing", there were search aggregators that simply returned results from a number of different search engines. This is the same but for the "AI".
looks like it could speed up my workflow
it looks quite useful, I'll try it.
Thanks Recep, we will do our best to make it even more useful.
hmm might try the free version see how it goes
This is insane!
Thanks!
thank you Keith!
How will this tool be affected by Apple's forthcoming AI in the next release of mac os?
Try the code editor cursor its a ai version of vscode
Typingmind is also good
Loved the term: Productainment !
that is good alex thanks
it will help me a lot😍😍
Thx!
You and Marques are probably the only ones that I find useful and genuine.
Brilliant product. I hope it sticks around, this was my side project list especially if they can get that local LLM feature the way I think it'll go.
Imagine it did an image search using google image search API then remix the results
I bought an average membership and start using it, it is much more useful and helpful than I expected
I running LLama 3 with Ollama for free. Just need apple silicon with at least 16gb ram.
I used it upon recommendation and it exceeded all my expectations.The Remix Feature allowed me to Compare it with the language models I want
Great! I need to try it
seems like itd be good for creative projects
Thank You 'Buddy' !
bro they have lifetime acceess option how is that possible?
Looks like a nice tool, thanks. Will give it a try. 👍
multiple ai models is pretty sweet ngl
The screen capture feature is my favorite😊
Wow. thank you for sharing. this is an amazing product that I need.
Thanks, Ismail. We will do our best to build the best LLM experience, enhanced by combining all LLMs in a single answer.
why most of the latest apps or software are starting only for macos
they do know that windows is still exist
I can't buy cos I'm Windows user
windows is for gaming, use mac for workinig
So is there any alternatives for Linux or a Web based app that is similar to thinkbuddy?
Produtainment? Genious!!!!!
Just use Cursor
The big missing feature here is full IDE integration so it can analyze the whole project / solution and propose something more than just single file / class refactoring or modification
I've tried several supposed IDE capable AI tools. So far, none of them can actually analyze my entire project. Even though the landing pages specifically say that it can.
remix thing looks neat
GitHub copilot left the chat
@@srijan_srivastava Copilot sucks ngl
hotkeys seem cool i guess
If AI features don’t come to europe witn apple’s next OS it might be a very good workaround and good deal
Mac : yes, iOS/iPadOS : maybe
I signed up from your link but missed the code. The discount doesn't matter for me. I just wish you had gotten credit.
No worries Amon, we will contact you and refund the partial amount to make it like you used the code, no worries!
Sorry if I am a Luddite, but do these send to the cloud? Is your data used for machine learning? I read that Samsung Fabs input their proprietary designs in one of these and gave the world access; not that anyone cared, as I assume they uploaded it because it needed help!
Or is it running on local LLMs?
support answered to me, that firstly our requirest are going to their api and only after that it goes to service providers. thus its not very nice.
Alternative for windows?
Hi there, this is out of topic but I wanted to ask you if you can use the ANE of the the m3 series chips for any type of ml training, I’m working in a segmentation model and I was wondering if I could use the ANE for training it, I’m now working with google colab and tensorflow
Bought lifetime subscription with your code
25 a month ain't bad if it does all that 👊🏻
Thanks for the 30% discount!!!
first human, no bot
Pretty speedy. Hello human. Bots were banned, but they keep on coming.
Do you have anything on Codeium? Thanks.
Alright then!
nice
Is it Mac app only, can we use on iPhone and Win (VM)?
We are preparing a new version with Phind/Perplexity-like web search features with more agentic features for all platforms, including Windows, Linux, Android, iOS, and a web version.
thought u were gonna compare some models with coding problems :(
wait he uses ai for emails? thats kinda interestng
Apple is going to steal their thunder, they are most likely going to give similar service in some time.
They were planning to give the AI tools on iPhones presently i believe, they will make it rock solid and look at the market for the demand (that these tools will create) and then crush all these tools.
amaziiiinggg
Maybe ill try if ppl say its good
So you waited a months to tell us?!!! We lost the time advantage now . . .
had to take it out for a good spin
It is not Alex's mistake, we just want to develop the world's most powerful LLM answers by remixing all powerful LLMs to compete with giants 😬
We have to wait a bit, your valuable insight will enlighten our roadmap. Please don't hesitate to share them with us.
Thanks for the reply Yiğit.
No worries though, I was just joking with Alex. He is very trusted source that I can count on.
Me, I am still looking for local LLMs solution. To extension to just text based LLMs, I mostly look to generate game objects (images, textures, ideally simple 3D models, etc).
Wish thinkbuddy great success.
@@DS-pk4eh Count us in, we are working on a new version that supports Local LLMs too, will happy to have your feedback in here.
Which mouse are you using alex ?
it’s a razer pro click mini
25$ a month for all?? No way dude!
Nice! I'll give it a try. Now let local LLMs hit the Snapdragon NPUs ;-) The non plus ultra having it all remixed together with Ollama/OpenWebUi, LM Studio....
We are going to integrate this feature into local LLMs soon and finally we can expect even better answers than Sonnet and GPT-4 for tasks that require less reasoning. Will be happy to hear feedback from local LLM evangelists too.
Selam Alex
Your promo code doesn't work
only for plus package I think
Look at cody
that remix thing seems neat wonder how it works
Jump onboard and make a SPONSORED VIDEO!!! jajajajaja
woah this thinkbuddy looks pretty cool
I tired it and mmmmg’b the interface and UX is horrible . First it’s not a macOS window viewport and it’s lacking features like a proper full screen . You can’t use it I. A white background, the glass bageok’d is distracting , no way to easily see past prompts on a side bar , impossible to drag document , bad use of the windows width .
alex mmmmh there’s no productivity enhancement aside for very short prompts requests . It’s not solid as a friendly daily user use
so what do you like about it
@@AZisk well it does the job if it’s a very short request and the remix is fine . But there is so much friction with the interface and when you have a bad sight like I do , it’s hard to focus with the glass effect , dark background and no font scaling .
Is the friction worth the price , I lm not sure yet though .
The lifetime license is tempting , but can it be trusted ? The lack of update , reviews , community around the project .
@@Gael_AGhaving the interface be customizable would be a great feature for them to add. good feedback!
@@AZisk the mandatory change should be a a proper mac window viewport with « traffic light « buttons . It’s the bare minimum
Hey Alex you do have a lot of MacBooks can you please exchange one with me? I bought an base M1 air 2 weeks ago (I have everything bill,box, purchase receipt) but it isn't sufficient for development task and here in India macbooks are just too costly and spec options aren't available. It would be of great help!
Time to be a member
raycast pro has this as well
I think this is so sus.. No one is reviewing it on youtube, only you.. their youtube channel is so sus, only 127 subscribers??
that’s what happens when you have a startup.
@@AZisk ok, so, you just replaced all AI tools for thinkbuddy? It uses exactly chat gpt 4o LLM or is it an approximation?
So if I get this right, we use a pile of crap to cover a pile of crap
kinda wanna try that code refactor thing 💪👏👏👏
ho
oOo :D
The remix feature makes my job much easier
Honestly, I hate AI times 1000!!!! It’s lame af
man waves closed fist at kids
Get the lfetime offer, I did and it's so worth it
Is vision actually unlimited ?
What is the context window ?
@@Gael_AG unlimited for lifetime users and plus monthly
hotkeys seem cool i guess