Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

NanoChat – The best ChatGPT that $100 can buy

github.com

Post reply on HN
Print view View original

321–326 of 326 posts

Re: NanoChat – The best ChatGPT that $100 can buy

#321

Post by AlexeyBelov »

kasey_junk wrote: post #245
troupo wrote: post #227

Earlier quoted context omitted.

Indeed. And only Karpathy is entitled to say that AI tools produce wrong code for him. And he's only entitled to say it for this project only. If anyone else says this, "the skepticism is exhausting", and their experience is completely irrelevant.

Go look at the comments on HN whenever someone posts about their AI coding workflow. It will be littered with negative comments that either imply or outright say that the poster is either shilling, ignorant or working only on toy examples. The grievance attitude seems to exist in both directions and is actually what is exhausting.

Show HNs about AI startups are littered with positive comments though. It's rare to see people calling out the submitter.

Re: NanoChat – The best ChatGPT that $100 can buy

#322

Post by megadragon9 »

Love the educational value of this "nano-sized" project. This reminded me of the from-scratch project I created to learn about deep learning libraries, neural networks all the way to LLMs like GPT-2 using just Numpy and Python [1]. Learning is done by "re-inventing the wheel" yourself, one step at a time :)

[1] https://github.com/workofart/ml-by-hand

Re: NanoChat – The best ChatGPT that $100 can buy

#323

Post by Breza »

sieve wrote: post #73

Nice! His Shakespeare generator was one of the first projects I tried after ollama. The goal was to understand what LLMs were about. I have been on an LLM binge this last week or so trying to build a from-scratch training and inference system with two back ends: - CPU (backed by JAX) - GPU (backed by wgpu-py). This is critical for me as I am unwilling to deal with the nonsense that is rocm/pytorch. Vulkan works for m…

What are your thoughts on using JAX? I've used TensorFlow and Pytorch and I feel like I'm missing out by not having experience with JAX. But at the same time, I'm not sure what the advantages are.

Re: NanoChat – The best ChatGPT that $100 can buy

#324

Post by sieve »

Breza wrote: post #323
sieve wrote: post #73

Nice! His Shakespeare generator was one of the first projects I tried after ollama. The goal was to understand what LLMs were about. I have been on an LLM binge this last week or so trying to build a from-scratch training and inference system with two back ends: - CPU (backed by JAX) - GPU (backed by wgpu-py). This is critical for me as I am unwilling to deal with the nonsense that is rocm/pytorch. Vulkan works for m…

What are your thoughts on using JAX? I've used TensorFlow and Pytorch and I feel like I'm missing out by not having experience with JAX. But at the same time, I'm not sure what the advantages are.

I only used it to build the CPU back end. It was a fair bit faster than the previous numpy back end. One good thing about JAX (unlike numpy) is that it also gives you access to a GPU back end if you have the appropriate stuff installed.

Re: NanoChat – The best ChatGPT that $100 can buy

#325

Post by llbbdd »

bgwalter wrote: post #306
llbbdd wrote: post #296

Earlier quoted context omitted.

When people take the time to virtue signal "Open" "AI" or the annoyingly common M$ on here I wonder often why they are wasting their precious time on earth doing that

It is in my style guide. I see that you optimized your time by omitting the full stop at the end of the sentence.

Totally fine, you're not alone. The kids call it a "typing quirk" as a justification if you're interested in learning more.

Re: NanoChat – The best ChatGPT that $100 can buy

#326

Post by spacecadet »

simonw wrote: post #276
spacecadet wrote: post #255

Built so many nano AIs over the last several years. I have played with nanoGPT, its ok. Just hype for Kpathy... So many tiny LLMs out there now that run on cheap SOCs. Try SmolVLM512, runs fine on a sub $100 pi.

You're misunderstanding the project. This isn't about an LLM that runs on $100 hardware. It's about a usable LLM that costs $100 to train from scratch.

No I get that. Having trained my own small LLMs and for much less than $100.
Post reply on HN