9523 MT/s is nice and all, but it's 2800 bucks for soldered only 32 GB RAM.. One can get a used/refurb 4060 Laptop for 800 (I get that this is not a gaming laptop) and it will be faster in games and the 5600 MT/s RAM will at least be upgradable to up to 64 GB and the 8 GB VRAM will allow for offloading AI LLM models partially or fully to.

Running AI LLM models locally and privately:
But why would one need more than 32 GB RAM you may ask? 32 GB RAM allow only a limited context of ~32000 tokens (= ~24000 words) for a quant of the new 35B MoE SOTA LLM Qwen3.6-35B-A3B-UD-Q4_K_M. And agentic workflows often require more than 100000 tokens, so it will not work.

Quote from: reddit.com/r/LocalLLaMA/comments/1sq94qx/is_anyone_getting_real_coding_work_done_with.. I've come to the conclusion that (1) 32768 is the biggest context I can get away with in an adequately smart model, and (2) it just ain't enough.

You praise Emma45, the AI, for wishful thinking. We have to await whether next CPUs will appear in 2027 or be further delayed.

@Enma45
Great post! :) I wanted to buy something like PL this year, but I will wait one more year for next gen CPUs and probably from a different manufacturer, than Intel.

At this price, almost $2,500, Intel won't sell anything in Europe. It's an insane price for this processor and such poor performance. Everyone is waiting for 2027 to see AMD Zen 6 and the successor to PantherLake, because by then there will be five competitors: Intel, AMD, Qualcomm, Nvidia, and Apple, and they'll have to lower their prices if they want to sell.