221–230 of 519 posts
Earlier quoted context omitted.
[flagged]
It may be more productive to ask what is right with burning fossil fuels for electricity right in the middle of marginalized communities that have to bear the cost of this pollution for AI slop.
Righteo, I guess I better suspend my white privilege.
"use all of the compute capacity at their Colossus 1 data center" So, they handed out all of their data center to Anthropic; Grok wasn't using it much?
> 300 megawatts of new capacity (over 220,000 NVIDIA GPUs) The scale is just mindboggling here. Are there any blog posts or anything discussing what kind of infrastructure is used for even just the inference side (nevermind the training) for SotA models like Opus? I would have thought it might be secret, but given that you can actually run the models yourself on AWS Bedrock doesn't that give an indication?
Earlier quoted context omitted.
Pretty smart for SpaceX though. They’re turning an asset they made for a money-pit (Grok) into probably a major source of revenue ahead of their IPO.
We all remember 2 weeks ago when SpaceX bought $10B of Cursor services. https://news.ycombinator.com/item?id=47855293 Since Cursor often relies on Claude models, some of those services will flow back to their own datacenter compute. Especially if there's, lets call it, "customer demand loadbalancing optimization agreements" that makes those Cursor services prioritize Claude models using the app keys that get load-bal…
Earlier quoted context omitted.
Pretty smart for SpaceX though. They’re turning an asset they made for a money-pit (Grok) into probably a major source of revenue ahead of their IPO.
I see it more of lets make money off the hardware we are not using anymore. From Elon on X: ... After that, I was ok leasing Colossus 1 to Anthropic, as SpaceXAI had already moved training to Colossus 2. https://x.com/elonmusk/status/2052069691372478511
Earlier quoted context omitted.
We all remember 2 weeks ago when SpaceX bought $10B of Cursor services. https://news.ycombinator.com/item?id=47855293 Since Cursor often relies on Claude models, some of those services will flow back to their own datacenter compute. Especially if there's, lets call it, "customer demand loadbalancing optimization agreements" that makes those Cursor services prioritize Claude models using the app keys that get load-bal…
I don't think it's the conspiracy theory that you're making it out to be. It is publicly known that the vast majority of deals in the AI space are circular in nature without the need for explicitly encoding any of it in a legal contract or even tacit agreements. e.g. Nvidia has invested significantly in many AI companies including both Anthropic and OpenAI which rely heavily on Nvidia's hardware and will undoubtedly…
Doubling the five-hour rate limits is merely a marketing stunt if the weekly rates are not also doubled. It simply means that you can reach the weekly limits in three days instead of five.
I don’t think I’ve hit either limit a single time in the past 5 months after upgrading to the $100 plan. On heavy weeks I probably am using it consistently for at least 6+ hours a day. Although, I’m pretty rigorous about always keeping my sessions under 200-250k tokens.
> 300 megawatts of new capacity (over 220,000 NVIDIA GPUs) The scale is just mindboggling here. Are there any blog posts or anything discussing what kind of infrastructure is used for even just the inference side (nevermind the training) for SotA models like Opus? I would have thought it might be secret, but given that you can actually run the models yourself on AWS Bedrock doesn't that give an indication?
> It’s regulation with the utilities. There are ramp rates, there are all of these things that you’re supposed to do to not screw up the grid. Data centers have been in gross violation of that. When you think about what’s wrong with data centers, they have load volatility, which we just talked about, then they decide to power it with behind-the-meter natural gas generators. These natural gas generators, their shaft is supposed to last for seven years. It’s lasting 10 months because of all the cycling.
https://www.volts.wtf/p/doing-data-centers-the-not-dumb-way
On the compute infrastructure, there are standard NVIDIA reference designs like this:
https://www.nvidia.com/en-us/technologies/enterprise-referen...
I haven't bothered to look but I'd guess Mellanox GPU-to-GPU networks, and massive custom code for splitting tensors across GPUs, and for shuttling activations across GPU nodes.
Earlier quoted context omitted.
As I understand it, the problem is cooling. There isn't any medium to take away the heat, so the only option is to slowly radiate it away.
Anyone who has googled just once to ask if datacenters in space make any sense, has found out they don't because they can't get rid of heat. That leaves only two kinds of people left who are still talking excitedly about datacenters in space: The uninformed and the grifters.
There’s very little research work needed to make this happen; it’s all about engineering some satellite buses and having them fly in close formation to get a “data center”. And this group of satellites in sun-synchronous orbit would relay to a comms constellation e.g. starlink itself) and operate as a global scale data center. The heat management and orbital mechanics are all straight forward really.