Ryan D. Hatch

a16z's Moses Sternstein and Alex Danco on the "rack spread," the gap between rising GPU prices and falling token prices: "The core observation in the rack spread is as follows: roughly since the summer, GPU pricing has gone up, token pricing has gone down. Why that is: a combination of open competition, the labs themselves cutting prices." "We start with a bunch of GPUs and a bunch of electricity and a bunch of cooling, and on the other side out comes tokens, right? And the thing that achieves that is the model." "In an oil refinery, you have a bunch of inputs... It goes into a refinery, which is like the model, and then out come refined products that you can actually use, like jet fuel or diesel or lubricant or any number of specialized things." "There is this term called the crack spread... It's not a net margin that anybody is actually making... But it is useful because if you look at any given refinery, just like you might look at any given model, you want to ask this basic question which is, 'Is this making any money?'" @MosesSternstein @alex_danco

Introducing SKY-001, the first Engineering World Model. On CAD and BIM tasks, it beats GPT-6 Astra and every frontier model we tested: -> 51-125x faster - > 150-210x cheaper -> Higher IOU than all frontier models SKY-001 lets you dream a physical object before you build it. 18 months of work. Built by the team out of Maluuba's deep learning lab (acq. by @MSRResearch). The first ‘AI General Engineer’ will be the backbone for Autonomous Manufacturing.

Marvell’s Dave Lazovsky is holding a tiny module that uses photonics to move high-bandwidth memory (HBM + DDR) as far as 50 meters away from the GPU. this can sit up to 50 meters from a GPU instead of the usual 1.5 mm. For the first time, memory no longer has to live right next to compute, letting AI clusters share massive capacity without being locked to the accelerator package. Latency stays in the hundreds of nanoseconds, close enough for practical memory use. Basically, Memory and compute now scale on separate axes instead of being locked together.

Meet Baby Samuel. He was aborted at 16 weeks and survived for 30 minutes, sucking his thumb for comfort. “No-one picked him up, wrapped him or gave him any medical care to alleviate his pain and suffering.” No child should be aborted.

We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks: Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better: Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better: Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better: Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work! In summary: - As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding. - Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.