Top Tweets for #B3IQ

CME planned to launch futures on GPU rental prices on October 5, but the CFTC extended its review through November 9. The issue is that there still isn’t a trusted market price for GPU compute. Most pricing happens in private deals, making it difficult to build a reliable benchmark for buyers trying to hedge future costs. The fact that companies increasingly need financial products just to manage the price of compute says a lot about where the market is heading. For teams that need reliable access and predictable economics, there’s another option: own the underlying compute instead of continually buying exposure to the rental market. https://t.co/Aorfjyu1XS

what one neocloud raised on oct 1: 1. RAM: +41% 2. AMD CPU cores: +25% 3. B300: +21% 4. H200: +20% 5. B200: +19% 6. H100: +17% that's nebius's on-demand price list. i flagged the CPU rise in september. what i missed: the biggest rise on the list is plain memory, ahead of every GPU. separately, on wednesday AWS raises its reserved GPU blocks 15% across B300, B200, H200 and H100. so the memory squeeze is now showing up on a cloud price list. trendforce sees DRAM contract prices up another 10-15% this quarter, so i don't think the RAM line is done. imo this is what a broad shortage of chips looks like. $NBIS

base:0xb3b32f9f8827d4634fe7d973fa1034ec9fddb3b3 was up 40%+ in the final week of September, wouldn’t complain about another week like that lol B3IQ is building a neocloud where you own the NVIDIA hardware, with the option to have it built and hosted in Oregon You decide whether it runs your own AI workloads or makes capacity available to others I’d like to see the GPU usage numbers alongside that chart

In 2009, NVIDIA shipped about 100 million GPUs a year. A couple of doctors at Mass General brought Jensen Huang a project where the best case was a research paper and an order for two or three GPUs. He said yes. In this interview from that year, Huang walks through how prioritizing that small buyer led to a breakthrough in medical imaging and net-new demand for NVIDIA GPUs across the medical industry. Buyers that size are often overlooked, but at B3 we see both the economic and societal potential of building for them. Early owners of B3IQ include faculty, researchers, and student teams at universities who run private, unmetered inference on dedicated GPU systems they own.

We compared renting GPUs in the cloud with owning a B3IQ system. Renting works for short bursts. An owned system handles those too, at a predictable cost, with your data on hardware nobody else shares. Andromeda assembles our systems in the U.S. Owners can rent out idle time as well. It's optional, and what it pays depends on demand and program terms.