Qualcomm's new chips stream 30B MoE weights from flash to RAM on demand
ryanshrout · x · 2026-09-23
Qualcomm staff walked through the new Snapdragon silicon: CPU, GPU, NPU, cache and a package-stacking design that saves space and improves thermals. The flagship technical feature is support for 30B MoE models, streaming weights from flash memory into system RAM on demand. GPU performance also improved; real-world perf remains to be seen.
Related event: Snapdragon 8 Elite Gen 6 brings 30B MoE models on-device(3 posts)→
More from Infra
- $500 of Dell OptiPlexes become a diskless netboot lab where AI agents can't brick the hardware — colinmcnamara · 2026-09-23
- AMD MI355X beats NVIDIA B200 by 2.25x at scale with ~40% lower cost per GPU hour: Signal65 — ryanshrout · 2026-09-23
- 2.3B MoE hybrid Mamba model matches Llama-3.2-3B with <1% of its pretraining FLOPs — tri_dao · 2026-09-23
- ByteDance Seed fixes full-pipeline FP8 RL instability with Calibrated Clipping across 8B-32B models — ByteDance-Seed · 2026-09-23
- DigitalOcean Managed Agents Enters Public Preview: Pause-When-Idle Cloud Claude Code and Codex — _AustinCalvert_ · 2026-09-23
- ARK analyst: AI is the most powerful joule in history, converting energy to GDP ~10x better than humans — DMaguireARK · 2026-09-23