#k80-ultra

#k80-ultra

"They have been more successful than anyone including themselves could have anticipated," said Nick Giles, a senior research analyst at B. Riley Securities who covers several of the firms, most of which are public.

gamers are probably going to feel left out since Nvidia seems to have decided renting cloud rigs to them is better than selling consumer hardware, small companies looking for AI chip compromises will be excited, and agentic AI is gonna be so hot that our Mann on the ground this week in San Jose isn't gonna need a jacket.

Claws are autonomous agents that can plan, act, execute tasks on their own, and they've gone from just thinking and executing on tasks to achieving entire missions. We used to prompt with what, how, or why, but for claws now we prompt with build, create or make.

Either way, I think the AI boom is alive and well, but with much of the short-term hype fading away, the big question is whether the long-term trajectory is still there and whether it makes sense for investors to hit the buy button now that the near-term is somewhat less hyped while the long-term is as exciting as ever.

AMD clarified those estimates are based on a comparison between an eight-GPU MI300X node and an MI500 rack system with an unspecified number of GPUs. The math works out to eight MI300Xs that are 1000x less powerful than X-number of MI500Xs. And since we know essentially nothing about the chip besides that it'll ship in 2027, pair TSMC's 2nm process tech with AMD's CDNA 6 compute architecture, and use HBM4e memory, we can't even begin to estimate what that 1000x claim actually means.

The new capabilities center on two integrated components: the Dynamo Planner Profiler and the SLO-based Dynamo Planner. These tools work together to solve the "rate matching" challenge in disaggregated serving. The teams use this term when they split inference workloads. They separate prefill operations, which process the input context, from decode operations that generate output tokens. These tasks run on different GPU pools. Without the right tools, teams spend a lot of time determining the optimal GPU allocation for these phases.

What we are beginning to see is that many open-source and other models are becoming compact enough and sufficiently optimized to run very efficiently on consumer GPUs,