Analyst: NAND's DRAM moment — DeepSeek V4.1 Flash keeps 26% of params on SSD
casper_hansen_ · x · 2026-09-11
Analyst Casper Hansen argues NAND is about to have its DRAM moment: in-memory parameter scaling is saturating as HBM capacity stops scaling massively.
- DeepSeek V4.1 Flash already keeps 26% of its parameters on SSD — "that's the new architecture"
- Storage is becoming part of the model itself
- His prediction: within two years, most parameters in frontier models won't be in memory at all
More from Infra
- Resin lets an agent build standalone LLM inference engines; vLLM and SGLang optimize DeepSeek kernels differently — AccBalanced · 2026-10-27
- Local benchmark: Flint matches Qwen3.6 quality at 50/52 while running 3.6x faster on M4 Pro — RichardSocher · 2026-10-27
- Anthropic S-1: CSP marketplace revenue jumped from 11% to 48% in three years — AccBalanced · 2026-10-27
- Alchemy adds effect/Workflow support across Durable Objects, Workflows and Lambda Durable Functions — samgoodwin89 · 2026-10-27
- Alchemy's Lambda Web Functions run plain Node web servers on Lambda, no OpenNext needed — samgoodwin89 · 2026-10-27
- Alchemy Cuts Cloudflare Bindings CPU Time by Up to 166× Via Profiling — samgoodwin89 · 2026-10-27