memossy

It was interesting Aurora used GPU Max & definitely looking forward to Falcon Shores.

I think Gaudi2 was bad timed & they had to build stack, Gaudi3 is where I think we will see mass adoption given availability, way cheaper price/performance & maturer stack.

There is still weird stuff when using them but they are surprisingly solid.

It's a great company & will do well, plenty of demand & B100s/BH200s etc coming

The Hopper stuff is particulalry interesting

Falcon shores next year will be crazy with 300gb VRAM & new lith

Gaudi2s started coming out in 2022 (https://huggingface.co/blog/habana-gaudi-2-benchmark) but didn't hit mass scale. I think Gaudi3 will & others have seen similar performance for Gaudi2 eg Databricks: https://www.databricks.com/blog/llm-training-and-inference-i..., mlperf etc

We are about to drop stable diffusion 3 which is the best image model out there (https://x.com/EMostaque/status/1764941367682256950?s=20) with similar architecture to Sora by OpenAI that can be used for any modality.

We have hundreds of millions of downloads of our models so are looking for big scale as we move to every pixel being generated & this stuff goes from research to mass deployment.

We use v4s, v5es & v5ps. Mostly v5ps, very stable int8 training (versus the horror that is fp8 stability)

I mean they work well, here is another blog by Databricks: https://www.databricks.com/blog/llm-training-and-inference-i...

I think as we go to enterprise workloads the total cost of ownership becomes important.

NVIDIA is still the best for research given ecosystem but once the models are standardised as with transformers/LLaMA and likely multimodal diffusion transformers it then becomes about scale, availability and cost per flop.

It took less than a day to port our code over, we do custom CUDA across modalities.

Gaudi2 was actually announced 2 years ago and is 7nm like the A100 80Gb it was meant to be competitive with, Gaudi3 later this year is probably going to be the inflection point as that ramps

The cost is like 1/3

https://www.intel.com/content/www/us/en/newsroom/news/vision...

The v5es and v5ps are pretty amazing at running SD, giving code for SD3 now to optimise it on those.

v5es are particularly interesting given the millions that will land and the large pod sizes, particularly well constructed for million token context windows.

Think it'll probably crack on with Gaudi3 at 4x performance, twice VRAM etc later this year.

We found cuda sycl conversion surprisingly good https://www.intel.com/content/www/us/en/developer/articles/t...

"For Stable Diffusion 3, we measured the training throughput for the 2B Multimodal Diffusion Transformer (MMDiT) architecture model. Gaudi 2 trained images 1.5x faster than the H100-80GB, and 3x faster than A100-80GB GPU’s when scaled up to 32 nodes. "

I could but I won't as legal stuff :)

I mean open models yo

If we trained it with videos yes but need more GPUs for that.

It'll be out soon, doing benchmark tests etc

We have highly efficient models for inference and a quantization team.

Need moar GPUs to do a video version of this model similar to Sora now they have proved that Diffusion Transformers can scale with latent patches (see stablevideo.com and our work on that model, currently best open video model).

We have 1/100th of the resources of OpenAI and 1/1000th of Google etc.

So we focus on great algorithms and community.

But now we need those GPUs.

As the leader in open image models it is incumbent upon us as the models get to this level of quality to take seriously how we can release open and safe models from a legal, societal and other considerations.

Not engaging in this will indeed lead to bad laws, sanctions and more as well as not fulfilling our societal obligations of ensuring this amazing technology is used for as positive outcomes as possible.

Stability AI was set up to build benchmark open models of all types in a proper way, this is why for example we are one of the only companies to offer opt out of datasets (stable cascade and SD3 are opted out), have given millions of supercompute hours in grants to safety related research and more.

Smaller players with less uptake and scrutiny don't need to worry so much about some of these complex issues, it is quite a lot to keep on top of, doing our best.

800m is good for mobile, 8b for graphics cards.

Bigger than that is also possible, not saturated yet but need more GPUs.

We did this for every stable diffusion release, you get the feedback data to improve it continuously ahead of open release.

Training on 4096 v5es how did you handle crazy batch size :o

You should go web3.

There are incredibly few lawyers who know anything about crypto and huge demand.

You can also join DAOs etc and earn significant amounts of money for your effort versus who you are.

If you want to do web3 code go to https://buildspace.so/ and you'll be in in a few months max.

I'm happy to have a chat about some of our web3 stuff too, semantic [at] gmail

"Payment can be accepted in any currency, including cryptocurrencies as we outrun the end of civilisation."

I was helped with my sinus pressure and dizziness by 2 x 1g N Acetyl Cysteine per day.

Very short half life, very quick response (if any), available on Amazon.

You may wish to look into Clemastine/Tavegil antihistamine, easy to get over the counter.

I would trust Tether more if it ever had meaningful outflows in dollar terms, had $1bn or so in the last crash (https://coinmarketcap.com/currencies/tether/) but has been generally on an upward march, not what you would expect from a stable coin

It's important to see if you are before or after the cliff.

If before then depending on your employment agreement and other docs there could be a scenario where you are fired/let go and get 0% shares.

Your last round valuation was $1,000,000 post so that price would be $141,000 or so for your 14% stake, can include some triggers on when that occurs that doesn't impede the business (ie $xm raised, $y profits).

If not then your ownership of the company is basically 10% on good leaver terms and that is the floor you should accept.

To illustrate assuming 100 total shares

Now: 40.8: him 39.2: you 10: option 10: seed investor

Goes to new cap table of: 40.8 him (57.6% ownership) 10 you (14% ownership) 10 option (14%) 10 seed investor (14%)

You're not going to get bought out now although you could say that your stake is purchaseable in the future at the last round valuation, which is very reasonable and keeps the cap table clear, probably $140,000 per the above with some sort of trigger for that (eg $xm raised, $y profits)

Lots of promise and a nice front end, but I'm struggling to see where after three years an example of high value knowledge in this platform is.

For example on the top topic AI most of the sections are unfilled / or lightweight aside from a list of companies with minimal contributions over the last few months: https://golden.com/wiki/Cluster%3A_Artificial_intelligence-J...

There is nothing on AI ethics (indeed this is the only thing on ethics: https://golden.com/wiki/Ethics-BJW8)

Looking at some comments above that maybe crypto was the angle I looked at a few articles and under Ethereum it says that the Constantinople release date was still to be determined (actual release date was 28th February 2019).

It seems there is a decent amount of company data in there (a la pitchbook, crunchbase), but in terms of practical, useful knowledge that is authoritative per the front page, what are some good examples now there?

I will try to get back to this thread with some resources but I would recommend as you are in the UK to look into clemastine/tavegil which you can get without prescription (noted in the article as about to be trialled with metformin) due to its action on P2X7 receptors/Schwann cells.

The ReBUILD doses were rather high though (10 tabs a day!). 1 tab would probably do it over a few months.

AI dungeon wrote an amazing Kafkaesque short story for me today based on just two tweets from some tech journos, just kept hitting enter:

https://twitter.com/EMostaque/status/1296411676749836288?s=2...

The microtransaction element doesn't require crypto.

This can be seen by the 0 fees on network on Alipay and WeChat Pay which don't use crypto and transact more volume now than Visa and Mastercard seamlessly.

With the rollout of DCEP why use the others?