Do these completely replace the nvidia drivers? Do they support DLSS 3 by chance? For some reason nvidia never released support for dlss 3/3.5 unless I missed it.
287 karma · joined December 22, 2023
Places like steam, discord, etc you very rarely see profile pictures above that size.
Some use cases might be generating profile pictures or banners for users or unique profile pictures for bots in online games. Discord, steam, social media, whatnot, you could just type what you want your profile picture to be and make it on the fly. They're small, aren't expected to be extremely high quality, and cheap enough.
Testing on https://fastsdxl.ai/ - "high quality profile picture of a cartoon cat holding a Bouquet of flowers"
To be clear it's not perfect, but this is a fairly complex prompt and I find the majority of seeds would be "good enough" for thumbnail profile pictures. I think we're almost there for "cheap good enough" usecases.
I'm not sure about re-doing the text of old images - you could try img2img but coherence is an issue, more controlnets might help
Machine learning, statistical models, procedural generation, literally an usage of heuristics are all being called "AI" nowadays which obfuscates the "boring" nature in favor of "exciting buzzword"
Selecting the quality of a video based on your download speed? That's "AI" now.
Very likely a coincidence.
https://opus-codec.org/release/stable/2024/03/04/libopus-1_5...
Their pricing suggests that either output tokens are more expensive for some technical reason, or they're trying to encourage a specific type of usage pattern, etc.
You're not wrong, but most of the big players will take a while to switch, at least in my experience you have to put more effort into making sure your prompts result in what you want, and that's annoying especially if GPT4 is already working for you. Claude historically has really bad refusals for safe prompts.
Also, GPT4 is cheaper 10/30 $/m vs 15/75 $/m for claude 3 opus - I'm not sure that price hike is worth the _slight_ benchmark improvement.
That suggests the inference time is more expensive then the memory needed to load it in the first place I guess?
The reason is the card transaction fees with most providers are usually a flat rate + %. $0.3+some% is typical I think?
And my last few month's invoices were 1.95, 3.81, 1.44, 2.09, 2.24 - those fees would suck at these prices.
I think if they had an auto-reload option it would fix the friction issue. Just a +5$ type of deal.
Well it being a stuttery mess for one. I could care less any of the features you mentioned when games or even the desktop is unusable.
Signal couldn't even render correctly.
I had a myriad of other issues, switching to x11 fixed it.
FWIW my previous experience a year ago was a stuttery mess making it a complete non-starter.
I do need to give it another go though.
I do need to give it another go though.
Of course mining is done on ASICs. If you care about latency you implement the network interface into the ASIC, and don't even use a CPU.
Parent's answer wasn't anti-crypto.
Depending on your definition, none of mistral's models are "open source", AFAIK they've given no information on what training data they used, how many tokens they trained it with, etc.
Is running an nvidia gpu + gnome/kde + steam viable these days? Do most games work well?
SDXL-ligtning/cascade can generate images in 200ms which is fast enough to fit in a web request, and paradoxically makes it even cheaper to generate.
And using groq at 500 t/s is wild compared to any of the other platforms.
All of the current frameworks support MoE and sharding among GPUs so I don't see what the issue is.
https://artificialanalysis.ai/models/mixtral-8x7b-instruct/h...
> On February 27, this feature will be start to be enabled automatically for all free accounts across GitHub.
https://github.blog/changelog/2024-02-14-secret-scannings-pu...
I don't think the naming really suggests that. The new naming suggests they'll have two sets, the "open" models and their commercial ones.
I do agree with your skepticism though. I kinda expected them to release something, likely an older model. Currently the closest is "miqu" which was a leak of a early quantized "medium".
The purpose of the "disorder" is to group common pathologies and their respective treatments (methods to improve).
Let's take high functioning autism as an example of a "expanding disorder". The "treatment" for high functioning autism is largely behavioural. Preparing for known stressful events, methods to stay focused on a task you don't like, practicing identifying and displaying emotions appropriately, etc.
> We could already improve ourselves without identifying as disordered.
Many people with high functioning autism don't see it as a disorder and view it as a "different thinking style".
We can (and I think should) "embrace" that, but that does nothing to improve difficulties such as dealing with a new situation, making friends, communicating with their boss, etc.
Similarly I've heard "everyone says that have trouble focusing, you don't have adhd".
Whether you have adhd or not, the ways to improve your behaviour are the same. Methods to stay organized, prepare your environment to focus, prioritizing, etc. apply regardless of whether "you really have adhd or not"
Hard disagree.
I do agree that for the majority of people, the 320kbit/s is absolutely enough for most people (arguably 160kbps is).
But it's not "hifi". And I don't mean "With extremely high end headphones and a dac, a trained ear, and holding your tongue at the right angle you can just barely make out a difference" - it's very noticable even with mediocre equipment. It's a rather lossy 320kbit/s too.
That's why a "hifi" tier makes sense, they don't currently serve that market at all.
I've seen a mix of both for years. For the last 6 years the story has usually been some combination of "X program technically supports it, but you'll get 20% of the performance you'd expect given the hardware and/or you'll need to go through a bunch of hoops, and even then you'll have to troubleshoot tons of random errors and/or it'll crash the program or system"
That's on top of the "nobody supports ROCm, or if they do it's because of a single person who amd the PR - you're on your own though because none of the core contributers have amd hardware" which I'll admit is a chick-egg problem.
But given these two factors, it's always meant "if you want buy hardware to do data science, you have to go nvidia if you don't want to write the support yourself, and/or an insane headache that often resulting in switching to nvidia hardware anyway"
It's possible they've sold it and we haven't heard about it since then. I would think they'd want to brag about it publicly though.
Reddit has a chicken and egg problem - nobody knows what the value of it's data even is. They've never sold it before. You can sell pet rocks for 1 billion dollars, but nobody will value them if there's been 0 sales. (ask vs bid vs last sale price).
I'm guessing reddit is rushing to sell their data so they can show that their data is worth anything at all.
And are there plans to release any more weights? Perhaps one or two revisions behind your latest ones?
Last I saw they performed really poorly, like lower single digits t/s. Don't get me wrong they're probably a decent value for experimenting with it, but is flat out pathetic compared to an A100 or H100. And I think useless for training?
The reason is because they instantly get a risk free guaranteed VERY healthy margin on every card they sell, and there's endless customers lined up for them.
If they kept the cards, they give up the opportunity to make those margins, and instead take the risk that they'll develop a money generating service (that makes more money then selling the cards).
This way there's no risk of: A competitor out competing them, not successfully developing a profitable product, "the ai bubble popping", stagnating development, etc.
There's also the advantage that this capital has allowed them to buy up most of TSMC's production capacity, which limits the competitors like Google's TPUs.
They are but it takes a lot of time.
Most of the big players - Google, Meta, OpenAI, Amazon, and Microsoft all are actively developing TPU/NPUs that would be used instead of the H100/A100's everyone is using for machine learning.
Google (tensorflow/jax), Meta(pytorch), Microsoft(onyx), Openai(triton) and Apple (mlx) each have software stacks for optimizing models for multiple platforms.
It takes a lot of time to develop the silicon and software stack. As a result everyone is using H100's in the interum until the hardware/software catches up. Google has been using their TPUs already.
There's other companies like Groq that are also developing NPU/TPU like devices.