projectmoon

Because of the porn or AI? 🙃

This is probably one of the best actual uses for something like generative AI. With enough data, they should be able to vectorize and translate dolphin language, assuming there is one.

That was entirely the point unfortunately.

Lol, there are smaller versions of Deepseek-r1. These aren’t the “real” Deepseek model, but they are distilled from other foundation models (Qwen2.5 and Llama3 in this case).

For the 671b parameter file, the medium-quality version weighs in at 404 GB. That means you need 404 GB of RAM/VRAM just to load the thing. Then you need preferably ALL of that in VRAM (i.e. GPU memory) to get it to generate anything fast.

For comparison, I have 16 GB of VRAM and 64 GB of RAM on my desktop. If I run the 70b parameter version of Llama3 at Q4 quant (medium quality-ish), it’s a 40 GB file. It’ll run, but mostly on the CPU. It generates ~0.85 tokens per second. So a good response will take 10-30 minutes. Which is fine if you have time to wait, but not if you want an immediate response. If I had two beefy GPUs with 24 GB VRAM each, that’d be 48 total GB and I could run the whole model in VRAM and it’d be very fast.

They’re probably referring to the 671b parameter version of deepseek. You can indeed self host it. But unless you’ve got a server rack full of data center class GPUs, you’ll probably set your house on fire before it generates a single token.

If you want a fully open source model, I recommend Qwen 2.5 or maybe deepseek v2. There’s also OLmo2, but I haven’t really tested it.

Mistral small 24b also just came out and is Apache licensed. That is something I’m testing now.

Most open/local models require a fraction of the resources of chatgpt. But they are usually not AS good in a general sense. But they often are good enough, and can sometimes surpass ChatGPT in specific domains.

Don’t know about “always.” In recent years, like the past 10 years, definitely. But I remember a time when Nvidia was the only reasonable recommendation for a graphics card on Linux, because Radeon was so bad. This was before Wayland, and probably even before AMD bought ATI. And it was certainly long before the amdgpu drivers existed.

For stuff like that, it’s best to have an auto formatter like checkstyle or something.

Had a team lead that kept requesting nitpicky changes, going in a FULL CIRCLE about what we should change or not, to the point that changes would take weeks to get merged. Then he had the gall to say that changes were taking too long to be merged and that we couldn’t just leave code lying around in PRs.

Jesus fucking Christ.

There’s a reason that team imploded…

Where can I get a sub 400 AMD card with 26 GB of VRAM?

https://agnos.is/posts/tech-recruitment-is-out-of-control.html

This was my experience at the beginning of 2024. It was bad enough that I had to write a blog post about it.

Have you tried Matrix?

LLMs are statistical word association machines. Or tokens more accurately. So if you tell it to not make mistakes, it’ll likely weight the output towards having validation, checks, etc. It might still produce silly output saying no mistakes were made despite having bugs or logic errors. But LLMs are just a tool! So use them for what they’re good at and can actually do, not what they themselves claim they can do lol.

Don’t think the snap is an official Mozilla package.

But wouldn’t you calculate the time in the future in the right time zone and then store it back as UTC?